NVMe: Four Key Trends Set to Drive Its Adoption in 2019 and Beyond

Storage vendors hype NVMe for good reason. It enables all-flash arrays (AFAs) to fully deliver on flash’s performance characteristics. Already NVMe serves as an interconnect between AFA controllers and their back end solid state drives (SSDs) to help these AFAs unlock more of the performance that flash offers. However, the real performance benefits that NVMe can deliver will be unlocked as a result of four key trends set to converge in the 2019/2020 time period. Combined, these will open the doors for many more companies to experience the full breadth of performance benefits that NVMe provides for a much wider swath of applications running in their environment.

Many individuals have heard about the performance benefits of NVMe. Using it, companies can reduce latency with response times measured in few hundred microseconds or less. Further, applications can leverage the many more channels that NVMe has to offer to drive throughput to hundreds of GBs per second and achieve millions of IOPs. These types of performance characteristics have many companies eagerly anticipating NVMe’s widespread availability.

To date, however, few companies have experienced the full breadth of performance characteristics that NVMe offers. This stems from:

  • The lack of AFAs on the market that fully support NVMe (about 20%).
  • The relatively small performance improvements that NVMe offers over existing SAS-attached solid-state drives (SSDs); and,
  • The high level of difficulty and cost associated with deploying NVMe in existing data centers.

This is poised to change in the next 12-24 months with four key trends poised to converge that will open up NVMe to a much wider audience.

  1. Large storage vendors getting ready to enter the NVMe market. AFA providers such as Tegile (Western Digital), iXsystems, Huawei, Lenovo, and others ship products that support NVMe. These vendors represent the leading edge of where NVMe innovation has occurred. However, their share of the storage market remains relatively small compared to providers such as Dell EMC, HPE, IBM, and NetApp. As these large storage providers enter the market with AFAs that support NVMe, expect market acceptance and adoption of NVMe to take off.
  2. The availability of native NVMe drivers on all major operating systems. The only two major enterprise operating systems that have currently native NVMe drivers for their OSes are Linux and VMware. However, until Microsoft and, to a lesser degree, Solaris, offer native NVMe drives, many companies will have to hold off on deploying NVMe in their environments. The good news is that all these major OS providers are actively working on NVMe drivers. Further, expect that the availability of these drivers will closely coincide with the availability of NVMe AFAs from the major storage providers and the release of the NVMe-oF TCP standard.
  3. NVMe-oF TCP protocol standard set to be finalized yet in 2018. Connecting the AFA controller to its backend SSDs via NVMe is only one half – and much easier part – of solving the performance problem. The much larger and more difficult problem is easily connecting hosts to AFAs over existing storage networks as it is currently difficult to setup and scale NVMe-oF. The establishment of the NVMe-oF TCP standard will remedy this and facilitate the introduction and use of NVMe-oF between hosts and AFAs using TCP/IP over existing Ethernet storage networks.
  4. The general availability of NVMe-oF TCP offload cards. To realize the full performance benefits of NVMe-oF using TCP, companies are advised to use NVMe-oF TCP offload cards. Using standard Ethernet cards with no offload engine, companies will still see high throughput but very high CPU utilization (up to 50 percent.) Using the forthcoming NVMe-oF TCP offload cards, performance increases by anywhere from 33 to 150 percent versus native TCP cards while only introducing nominal amounts of latency (single to double digit microseconds.)

The business need for NVMe technology is real. While today’s all-flash arrays have tremendously accelerated application performance, NVMe stands poised to unleash another round of up to 10x or more performance improvements. But to do that, a mix of technologies, standards, and programming changes to existing operating systems must converge for mass adoption in enterprises to occur. This combination of events seems poised to happen in the next 12-24 months.




Six Best Practices for Implementing All-flash Arrays

Almost any article published today related to enterprise data storage will talk about the benefits of flash memory. However, while many organizations now use flash in their enterprise, most are only now starting to use it at a scale where they use it to host more than a handful of their applications. As organizations look to deploy flash more broadly in their enterprises, here are six best practices to keep in mind as they do so.

The six best practices outlined below are united by a single overarching principle. That overarching principle is that the data center is not merely a collection of components, it is an interdependent system. Therefore, the results achieved by changing any key component will be constrained by its interactions with the performance limits of other components. Optimal results come from optimizing the data center as a system.

Photograph of scaffolding on a building

Photo by Dan Gold on Unsplash

Best Practice #1: Focus on Accelerating Applications

Business applications are the reason businesses run data centers. Therefore, accelerating applications is a useful focus in evaluating data center infrastructure investments. Eliminating storage perfor­mance bottlenecks by implementing an all-flash array (AFA) may reveal bottlenecks elsewhere in the infrastructure, including in the applications themselves.

Getting the maximum performance benefit from an AFA may require more or faster connections to the data center network, changes to how the network is structured and other network configuration details. Application servers may require new network adapters, more DRAM, adjustments to cache sizes and other server configuration details. Applications may require configuration changes or even some level of recoding. Some AFAs include utilities that will help identify the bottle­necks wherever they occur along the data path.

Best Practice #2: Mind the Failure Domain

Consolidation can yield dramatic savings, but it is prudent to consider the failure domain, and how much of an organization’s infrastructure should depend on any one component—including an all-flash array. While all the all-flash arrays that DCIG covers in its All-flash Array Buyer’s Guides are “highly available” by design, some are better suited to deliver high availability than others. Be sure the one you select matches your requirements and your data center design.

Best Practice #3: Use Quality of Service Features and Multi-tenancy to Consolidate Confidently

Quality of Service (QoS) features enable an array to give criti­cal business applications priority access to storage resources. Multi-tenancy allocates resources to specific business units and/or departments and limits the percentage of resources that they can consume on the all-flash array at one time. Together, these features protect the array from being monopolized by any one application or bad actor.

Best Practice #4: Pursue Automation

Automation can dramatically reduce the amount of time spent on routine storage management and enable new levels of IT agility. This is where features such as predictive analytics come into play. They help to remove the risk associated with managing all-flash arrays in complex, consolidated environments. For instance, they can proactively intervene by identifying problems before they impact production apps and take steps to resolve them.

Best Practice #5: Realign Roles and Responsibilities

Implementing an all-flash storage strategy involves more than technology. It can, and should, reshape roles and responsibilities within the central IT department and between central IT, develop­ers and business unit technologists. Thinking through the possible changes with the various stakeholders can reduce fear, eliminate obstacles, and uncover opportunities to create additional value for the business.

Best Practice #6: Conduct a Proof of Concept Implementation

A good proof-of-concept can validate feature claims and uncover perfor­mance-limiting bottlenecks elsewhere in the infrastructure. However, key to implementing a good proof-of-concept is having an environment where you can accurately host and test your production environment on the AFA.

A Systems Approach Will Yield the Best Result

Organizations that approach the AFA evaluation from a systems perspective–recognizing and honoring the fact that the data center is an interdependent system that includes hardware, software and people—and that apply these six best practices during an all-flash array purchase decision are far more likely to achieve the objectives that prompted them to look at all-flash arrays in the first place.

DCIG is preparing a series of all-flash array buyer’s guides that will help organizations considering the purchase of an all-flash array. DCIG buyer’s guides accelerate the evaluation process and facilitate better-informed decisions. Look for these buyer’s guides beginning in the second quarter of 2018. Visit the DCIG web site to discover more articles that provide actionable analysis for your data center infrastructure decisions.




Two Most Disruptive Storage Technologies at the NAB 2018 Show

The exhibit halls at the annual National Association of Broadcasters (NAB) show in Las Vegas always contain eye-popping displays highlighting recent technological advances as well as what is coming down the path in the world of media and entertainment. But behind NAB’s glitz and glamour lurks a hard, cold reality; every word recorded, every picture taken, and every scene filmed must be stored somewhere, usually multiple times, and available at a moment’s notice. It is these halls at the NAB show that DCIG visited where it identified two start-ups with storage technologies poised to disrupt business as usual.

Storbyte. Walking the floor at NAB, a tall, blond individual literally yanked me by the arm as I was walking by and asked me if I had ever heard of Storbyte. Truthfully, the answer was No. This person turned out to be Steve Groenke, Storbyte’s CEO, and what ensued was a great series of conversations while at NAB.

Storbyte has come to market with an all-flash array. However, it took a very different approach to solve the problems of longevity, availability and sustainable high write performance in SSDs and storage systems built with them. What makes it so disruptive is it created a product that meets the demand for extreme sustained write performance by slowing down flash and it does so at a fraction of the cost of what other all-flash arrays cost.

In looking at today’s all-flash designs, every flash vendor is actively pursuing high performance storage. The approach they take is to maximize the bandwidth to each SSD. This means their systems must use PCIe attached SSDs addressed via the new NVMe protocol.

Storbyte chose to tackle the problem differently. Its initial target customers had continuous, real-time capture and analysis requirements as they routinely burned through the most highly regarded enterprise class SSDs in about seven months. Two things killed NAND flash in these environments: heat and writes.

To address this problem, Storbyte reduces heat and the number of writes that each flash module experiences by incorporating sixteen mSATA SSDs into each of its Eco*Flash SSDs. Further, Storbyte slows down the CPUs in each of the mSATA module on its system and then wide-stripes writes across all of them. According to Storbyte, this only requires about 25% of the available CPU on each mSATA module so they use less power. By also managing the writes, Storbyte simultaneously extends the life of each mSATA module on its Eco-flash drives.

The end result is a low cost, high performance, very dense, power-efficient all-flash array built using flash cards that rely upon “older”, “slower”, consumer-grade mSATA flash memory modules that can drive 1.6 million IOPS on a 4U system. More notably, its systems cost about a quarter of that of competitive “high performance” all-flash arrays while packing more than a petabyte of raw flash memory capacity in 4U of rack space that use less power than almost any other all-flash array.

Wasabi. Greybeards in the storage world may recognize the Wasabi name as a provider of iSCSI SANs. Well, right name but different company. The new Wasabi recently came out of stealth mode as a low cost, high performance, cloud storage provider. By low cost, we mean 1/5 of the cost of Amazon’s slowest offering (Glacier) and at 6x the speed of Amazon’s highest performing S3 offering. In other words, you can have your low cost cloud storage and eat it too.

What makes its offering so compelling is that it offers storage capacity at $4.99/TB per month. That’s it. No additional egress charges for every time you download files. No complicated monthly statements to decipher to figure out how much you are spending and where. No costly storage architects to hire to figure out how to tier data to optimize performance and costs. This translates into one fast cloud storage tier at a much lower cost than the Big 3 (Amazon AWS, Google Cloud, and Microsoft Azure.)

Granted, Wasabi is a cloud storage provider start-up so there is an element of buyer beware. However, it is privately owned and well-funded. It is experiencing explosive growth with over 1600 customers in just its few months of operation. It anticipates raising another round of funding. It already has data centers scattered throughout the United States and around the world with more scheduled to open.

Even so, past horror stories about cloud providers shutting their doors give every company pause by using a relatively unknown quantity to store their data. In these cases, Wasabi recommends that companies use its solution as your secondary cloud.

Its cloud offering is fully S3 compatible and most companies want a cloud alternative anyway. In this instances, store copies of your data to both Amazon and Wasabi. Once stored, run any queries, production, etc. against the Wasabi cloud. The Amazon egress charges that your company avoids by accessing its data on the Wasabi cloud will more than justify taking the risk of storing the data you routinely access on Wasabi. Then in the unlikely event Wasabi does go out of business (not that it has any plans to do so,) companies still have a copy of data with Amazon that they can fail back to.

This argument seems to resonate well with prospects. While I could not substantiate these claims, Wasabi said that they are seeing multi-petabyte deals coming their way on the NAB show floor. By using Wasabi instead of Amazon in the use case just described, these companies can save hundreds of thousands of dollars per month just by avoiding Amazon’s egress charges while mitigating their risk associated with using a start-up cloud provider such as Wasabi.

Editor’s Note: The spelling of Storbyte was corrected on 4/24.




NVMe: Setting Realistic Expectations for 2018

Non-volatile Memory Express (NVMe) has captured the fancy of the enterprise storage world. Implementing NVMe on all-flash arrays or hyper-converged infrastructure appliances carries with it the promise that companies can leverage these solutions to achieve sub-millisecond response times, drive millions of IOPS, and deliver real-time application analytics and transaction processing. But differences persist between what NVMe promises for these solutions and what they can deliver. Here is a practical look at NVMe delivers on these solutions in 2018.

First and foremost, NVMe is an exciting and needed breakthrough to deliver on the performance characteristics as of early 2018. Unlike the SCSI protocol that it replaces which was designed and implemented with mechanical hard disk drives (HDDs) in mind, NVMe comes to market intended for use with today’s flash-based systems. In fact, as evidence of the biggest difference between SCSI and NVMe, NVMe cannot even interface with HDDs. NVMe is intended to speak flash.

As part of speaking flash, NVMe no longer concerns itself with the limitations of mechanical HDDs. By way of example, HDDs can only handle one command at a time. Whether it is a read or a write, the entire HDD is committed to completing that one command before it can start processing the next one and it only has one channel delivering that command to it.

The limitations of flash, and by extension, NVMe, are exponentially higher. In the case of NVMe, it can support 65,535 queues into the flash media and stack up to 64,000 commands per queue. In other words, over 4 billion commands can theoretically be issued to a single flash media at any time.

Of course, just because NVMe can support over 4 billion commands does not mean that any product or application currently even comes close to doing that. Should they ever do so, and they probably will at some point, it is plausible that published IOP numbers might be in the range of tens or hundreds of millions of IOPs. But as of early 2018, everyone must still develop and mature their infrastructure and applications to support that type of throughput. Further, NVMe as a protocol still must continue to mature its interface to support those kinds of workloads.

So as of early 2018, here is what enterprises can realistically expect from NVMe:

1. If you want NVMe on your all-flash array, you have a short list from which to choose. NVMe capable all-flash arrays that have NVMe interfaces to all SSDs are primarily available from Dell EMC, Huawei, Pivot3, Pure Storage, and Tegile. The number of all-flash arrays that currently support NVMe remains in the minority with only 18% of the 100+ all-flash arrays that DCIG evaluated supporting NVMe connectivity to all back end SSDs.

Source: DCIG

The majority of AFAs currently shipping support a 3, 6, or 12 Gb SAS interface to their backend flash media for good reason: few applications can take full advantage of NVMe’s capabilities. As both applications and NVMe mature, expect the number of AFAs that support NVMe to increase.

2. Your connectivity between your server and shared storage array will likely remain the same in 2018. Enterprises using NAS protocols such as CIFS or NFS or SAN protocols such as FC or iSCSI should expect to do so for 2018 and probably for the next few years. While new standards such as NVMe-oF are emerging and provide millions of IOPs when implemented, such as evidenced by early solutions from providers such as E8 Storage, NVMe is not yet well suited to act as a shared storage protocol between servers and AFA arrays. For now, NVMe remains best suited for communication between storage array controllers and their backend flash media or on servers that have internal flash drives. To use NVMe for any other use cases in enterprise environments is, at this point, premature.

3. NVMe is a better fit for hyper-converged infrastructure solutions than AFAs for now. Enterprises expecting a performance boost from their use of NVMe will likely see it whether they deploy it in hyper-converged infrastructure or AFA solutions. However, enterprises must connect to AFAs using existing storage protocols such as listed above. Conversely, applications running on hyper-converged infrastructure solutions that support NVMe may see better performance than those running on AFAs. Using AFAs, protocol translation over a NAS or SAN must still occur over the storage network to get to the NVMe enabled AFA. Hyper-converged infrastructure solutions negate the need for this additional protocol conversion.

NVMe will improve performance but verify your applications are ready. Stories about the performance improvements that NVMe offers are real and validated in the real world. However, these same users also find that some of their applications using these NVMe-based all-flash arrays are not getting the full benefit that they expected from them because, in part, their applications cannot handle the performance. Some users report that they have uncovered their applications have wait times built into them because the applications were designed to work with slower HDDs. Until the applications themselves are updated to account for AFAs by having those preconfigured wait times removed or minimized, the applications may become the new choke point that prevent enterprises from reaping the full performance benefits that NVMe has to offer.

NVMe is almost without doubt the future for communicating with flash media. But in early 2018, enterprises need to set realistic expectations as to how much of a performance boost NVMe will provide when deployed. Sub-millisecond response times are certainly a realistic expectation and maybe almost a necessity at this point to justify the added expense of using an NVMe array since many SAS-based arrays may achieve this same metric. Further, once an enterprise commits to using NVMe, one also makes the commitment to only using flash media since NVMe provides no option to interface with HDDs.




Data Center Efficiency, Performance, Scalability: How Dell EMC XtremIO, Pure Storage Flash Arrays Differ

Latest DCIG Pocket Analyst Report Compares Dell EMC XtremIO and Pure Storage All-flash Product Families

Hybrid and all-disk arrays still have their place in enterprise data centers but all-flash arrays are “where it’s at” when it comes to hosting and accelerating the performance of production applications. Once reserved only for applications that could cost-justify these arrays, continuing price erosion in the underlying flash media coupled with technologies such as compression and deduplication have put these arrays at a price point within reach of almost any size enterprise. As that occurs, flash arrays from Dell EMC XtremIO and Pure Storage are often on the buying short lists for many companies.

When looking at all-flash arrays, it is easy to fall into the trap that they are all created equal. While it can be truthfully said that every all-flash array is faster and will outperform any of its all-disk or hybrid storage array predecessors, there can be significant differences in how effectively and efficiently each one delivers that performance.

Consider product families from leaders in the all-flash array market: Dell EMC XtremIO and Pure Storage. When you look at their published performance specifications, they both scale to offer hundreds of thousands of IOPS, achieve sub one millisecond response times, and offer capacity optimization features such as compression and deduplication.

It is only when you start to pull back the covers on these two respective product lines that substantial differences between them start to emerge such as:

  • Their data center efficiency in areas such as power consumption and data center footprint
  • How much flash capacity they can ultimately hold
  • What storage protocols they support

This recent published 4-page DCIG Pocket Analyst Report analyzes these attributes and others on all-flash arrays from these two providers. It examines how well their features support these key data center considerations and includes analyst commentary on which product has the edge in this these specific areas. This report also contains a feature comparison matrix to support this analysis.

This report provides the key insight in a concise manner that enterprises need to make the right choice in an all-flash array solution for the rapidly emerging all-flash array data center. This report may be purchased for $19.95 at TechTrove, a new third-party site that hosts and makes independently developed analyst content available for sale.

All-flash data centers are coming and with every all-flash array providing higher levels of performance than previous generations of storage arrays, enterprises need to examine key underlying features that go deeper than simply fast they perform. Their underlying architecture, the storage protocols they support, and the software they use to deliver these features are all features that impact how effective and efficient the array will be in your environment. This DCIG Pocket Analyst Report makes plain some of the key ways that the all-flash arrays from Dell EMC and Pure Storage differentiate themselves from one another. Follow this link to purchase this report.

Author’s Note: The link to the DCIG Pocket Analyst Report comparing the Dell EMC XtremIO and Pure Storage FlashArrays was updated and correct at 12:40 pm CT on 10/18/2017 to point to the correct page on the TechTrove website. Sorry for any confusion!




Software-defined Data Centers Have Arrived – Sort of

Today organizations more so than ever are looking to move to software-defined data centers. Whether they adopt software-defined storage, networking, computing, servers, security, or all of them as part of this initiative, they are starting to conclude that a software-defined world trumps the existing hardware defined one. While I agree with this philosophy in principle, organizations need to carefully dip their toe into the software-defined waters and not dive head-first.

The concept of software-defined data centers is really nothing new. This topic has been discussed for decades and was the subject of one of the first articles I ever published 15 years ago (though the technology was more commonly called virtualization at that time.) What is new, however, is the fact that the complementary, supporting set of hardware technologies needed to enable the software-defined data center now exists.

More powerful processors, higher capacity memory, higher bandwidth networks, scale-out architectures, and other technologies have each contributed, in part, to making software-defined data centers a reality. The recent availability of solid state drives (SSDs) may have been perhaps the technology that ultimately enabled this concept to go from the drawing boards into production. SSDs reduce data access times from milliseconds to microseconds helping to remove one of the last remaining performance bottlenecks to making software-defined data centers a reality.

Yet as organizations look to replace their hardware defined infrastructure with a software-defined data center, they must still proceed carefully. Hardware defined infrastructures may currently cost a lot more than software-defined data centers but they do offer distinct benefits that software-defined solutions currently are still hard-pressed to match.

For instance, the vendors who offer the purpose-built appliances for applications, backup, networking, security, or storage used in hardware defined infrastructures typically provide hardware compatibility lists (HCLs). Each HCL names the applications, operating systems, firmware, etc., for which the appliance is certified to interact with and which the vendor will provide support. Deviate from that HCL and your ability to get support suddenly gets sketchy.

Even HCLs are problematic due to the impossibly large number of possible configurations that exist in enterprise environments which vendors can never thoroughly vet and test.

This has led to the emergence of converged infrastructures. Using these, vendors guarantee that all components in the stack (applications, servers, network, and storage along with their firmware and software) are tested and certified to work together. So long as organizations use the vendor approved and tested hardware and software component in this stack and keep them in sync with the vendor specifications, they should have a reliable solution.

Granted, obtaining solutions that satisfy these converged infrastructure requirements cost more. But for many enterprises paying the premium was worth it. This testing helps to eliminate situations such as I once experienced many years ago.

We discovered in the middle of a system wide SAN upgrade that a FC firmware driver on all the UNIX systems could not detect the LUNs on the new storage systems. Upgrading this driver required us to spend nearly two months with individuals coming in every weekend to apply this fix across all these servers before we could implement and use the new storage systems.

Software-defined data centers may still encounter these types of problems. Even though the software itself may work fine, it cannot account for all the hardware in the environment or guarantee interoperability with them. Further, since software-defined solutions tend to go into low cost and/or rapidly changing environments, there is a good possibility the HCLs and/or converged solutions they do offer are limited in their scope and may have not been subjected to the extensive testing that production environments.

The good news is that software-defined data centers are highly virtualized environments. As such, copies of production environments can be made and tested very quickly. This flexibility mitigates the dangers of creating unsupported, untested production environments. It also provides organizations an easier, faster means to failback to the original configuration should the configuration now work as expected.

But here’s the catch. While software-defined data centers provide flexibility, someone must still possess the skills and knowledge to make the copies, perform the tests, and do the failbacks and recoveries if necessary. Further, software-defined data centers eliminate neither their reliance on underlying hardware components nor the individuals who create and manage them.

Interoperability with the hardware is not a given and people are known to be unpredictable and/or unreliable from time to time, the whole system could go down or function unpredictably without a clear path to resolution. Further, if one encounters interoperability issues initially or at some point in the future, the situation may get thornier. Organizations may have to ask and answer questions such as:

  1. When the vendors start finger pointing, who owns the problem and who will fix it?
  2. What is the path to resolution?
  3. Who has tested the proposed solution?
  4. How do you back out if the proposed solution goes awry?

Software-defined data centers are rightfully creating a lot of buzz but they are still not the be all and end all. While the technology now exists at all levels of the data center to make it practical to deploy this architecture and for companies to realize significant hardware savings in their data center budgets, the underlying best practices and support needed to successfully implement software-defined data are still playing catch-up. Until those are fully in place or you have full assurances of support by a third party, organizations are advised to proceed with caution on any software-defined initiative, data center or otherwise.




Nimbus Data Reset Puts its ExaFlash D-Series at Forefront of All-flash Array Cost/Performance Curve

A few years ago when all-flash arrays (AFAs) were still gaining momentum, newcomers like Nimbus Data appeared poised to take the storage world by storm. But as the big boys of storage (Dell, HDS, and HPE, among others,) entered the AFA market, Nimbus opted to retrench and rethink the value proposition of its all-flash arrays. Its latest AFA models, the ExaFlash D-Series, is one of the outcomes of that repositioning as these arrays answer the call of today’s hosting providers. These arrays deliver the high levels of availability, flexibility, performance, and storage density that they seek backed by one of the lowest cost per GB price points in the market.

To get a better handle on the changes that have occurred at Nimbus Data over the past few years and the AFA market in general, I spoke with its CEO and founder, Thomas Isakovich. As the predominant enterprise storage players entered the AFA market, Nimbus had to first quantify the ways in which its models differentiated themselves from the pack and then communicate that message to the market place.

In comparing its features to its competitors, it identified some areas where its products out shined the competition. Specifically, its models offered support for multiple different high performance storage network protocols, it had a much lower price point on a per/TB basis, and its all-flash D-Series (one of its four platforms) pack much more flash into a 4U rack unit than models from the largest AFA providers. Notably, the research in the DCIG Competitive Intelligence Portal backs up these claims as the chart below reveals.

DCIG Comparison of Key Nimbus ExaFlash D-Series to Large AFA Providers

Source: DCIG Competitive Intelligence Portal; Names of Competitive Models Available with Paid Subscription*

This analysis of its product feature helped Nimbus to refine and better articulate its go-to-market strategy. For instance, thanks to its scale-out design coupled with its very high performance and low cost point, Nimbus now primarily focuses its sales efforts on hyper-scalers that need AFAs with these specific attributes.

Nimbus finds its most success with cloud infrastructure companies as well as organizations in the life sciences, post-production, and financial services markets. Further, due to the size and specific needs of customers in these markets, it suspended sales through the channel and has switched to primarily relying on direct sales.

The conversation I had with its CEO revealed that Nimbus is alive and well in the AFA market and still innovating much as it did when it first arrived on the scene years ago. However, it is also clear that Nimbus has a much better grasp of its competitive advantages in the market place and has adapted its go-to market plan based upon that insight to ensure its near and long term success.




Server-based Storage Makes Accelerating Application Performance Insanely Easy

In today’s enterprise data centers, when one thinks performance, one thinks flash. That’s great. But that thought process can lead organizations to think that “all-flash arrays” are the only option they have to get high levels of performance for their applications. That thinking is now so outdated. The latest server-based storage solution from Datrium illustrates how accelerating application performance just became insanely easy by simply clicking a button versus resorting to upgrading some hardware in their environment.

As flash transforms the demands of application owners, organizations want more options to cost-effectively deploy and manage it. These include:

  • Putting lower cost flash on servers as it performs better on servers than across a SAN.
  • Hyper-converged solutions have become an interesting approach to server-based storage. However, concerns remain about fixed compute/capacity scaling requirements and server hardware lock-in.
  • Array-based arrays have taken off in large part because they provide a pool of shared flash storage accessible to multiple servers.

Now a fourth, viable flash option has appeared on the market. While I have always had some doubts about server-based storage solutions that employ server-side software, today I changed my viewpoint after reviewing Datrium’s DVX Server-powered Storage System.

Datrium has the obvious advantages over arrays as it leverages the vast, affordable and often under-utilized server resources.  But unlike hyper-converged systems, it scales flexibly and does not require a material change in server sourcing.

To achieve this ends, Datrium has taken a very different approach with its “server-powered” storage system design.  In effect, Datrium split speed from durable capacity in a single end-to-end system.  Storage performance and data services tap host compute and flash cache, driven by Datrium software that is uploaded to the virtual host. It then employs its DVX appliance, an integrated external storage appliance, that permanently holds data and orchestrates the DVX system protects application data in the event of server or flash failure.

This approach has a couple meaningful takeaways versus traditional arrays:

  • Faster flash-based performance given it is local to the server versus accessed across a SAN
  • Lower cost since server flash drives cost far less than flash drives found on an all-flash array.

But it also addresses some concerns that have been raised about hyper-convered systems:

  • Organizations may independently scale compute and capacity
  • Plugs into an organization’s existing infrastructure.

Datrium Offers a New Server-based Storage Paradigm

StatelessServers_Diesl-1024x818

Source: Datrium

Datrium DVX provides the different approach needed to create a new storage paradigm. It opens new doors for organizations to:

  1. Leverage excess CPU cycles and flash capacity on ESX servers. ESX servers now exhibit the same characteristics that the physical servers they replaced once did: they have excess, idle CPU. By deploying server-based storage software at the hypervisor level, organizations can harness this excess, idle CPU to improve application performance.
  2. Capitalize on lower-cost server-based flash drives. Regardless of where flash drives reside (server-based or array-based,) they deliver high levels of performance. However, server-based flash costs much less than array-based flash while providing greater flexibility to add more capacity going forward.

Accelerating Application Performance Acceleration Just Became Insanely Easy

Access to excess server-based memory, CPU and flash combine to offer another feature that array-based flash can never deliver: push button application performance. By default, when the Datrium storage software installs on ESX hypervisor, it limits itself to 20 percent of the available vCPU available to each VM. However, not every VM uses all of its available vCPU with many VMs only using only 10-40 percent of their available resources.

Using Datrium’s DIESL Hyperdriver Software version 1.0.6.1, VM administrators can non-disruptively tap into these latent vCPU cycles. Using Datrium’s new Insane Mode, they may increase the available vCPU cycles a VM can access from 20 to 40 percent with a click of a button. While the host VM must have latent vCPU cycles available to accomplish this task, this is a feature that array-based flash would be hard-pressed to ever offer and unlikely could ever do with the click of a button.

Server-based storage designs have shown a lot of promise over the years but have not really had the infrastructure available to them to build a runway to success. That has essentially changed and Datrium is one of the first solutions to come to market that recognizes this fundamental change in the infrastructure of data centers and has brought a product to market to capitalize on it. As evidenced by the Insane Mode in its latest software release, organizations may now harness next generation server-based storage designs and accelerate application performance while dramatically lowering complexity and costs in their environment.




X-IO Refers to iglu as “Intelligent Storage”; I Say It is “Thoughtful”

In the last couple of weeks X-IO announced a number of improvements to its iglu line of storage arrays – namely flash optimized controllers and stretch clustering. But what struck me in listening to X-IO present the new features of this array was in how it kept referring to the iglu as “intelligent.” While that term may be accurate, when I look iglu’s architecture and data management features and consider them in light of what small and midsize enterprises need today, I see the iglu’s architecture as “thoughtful.”

Anyone familiar with the X-IO product line knows that it has its origins in hardware excellence. It was one of the first to offer a 5-year warranty for its arrays. It has been almost maniacal in its quest to drive every ounce of availability, performance and reliability out of the hard disk drives (HDDs) used in its systems. It designed its arrays in such a way that it made them very easy and practical to scale without having to sacrifice performance or manageability to do so.

But like many providers of HDD arrays, X-IO has seen its apple cart upset by the advent of flash. Hardware features once seen imperative or at least highly desirable on HDD-based arrays may now matter little or not at all. In some cases, they even impede flash’s adoption in the array.

X-IO arguably also encountered some of these same issues when flash first came out. To address this, its ISE and iglu arrays carried forward X-IO’s strengths of having an in-depth understanding of the media in its arrays. These arrays now leverage that to optimize the performance of flash media to drive up to 600,000 IOPS of performance on either its hybrid or all-flash arrays. The kicker, and what I consider to be the “thoughtful” part of X-IO’s array architecture, is in how it equips organizations to configure these two arrays.

home_page_product_ise1-300x149

Source: X-IO Technologies

One of the primary reasons behind flash’s rapid adoption has been its high levels of performance – up to 10x or more than that of HDDs. This huge performance boost has led many organizations to deploy flash to support their most demanding applications.

Yet this initial demand for performance by many organizations coupled with the need for storage providers to quickly deliver on this demand resulted in many all-flash arrays either lacking needed data management features (clustering, replication, snapshots, etc.) or delivering versions of these services that were really not enterprise ready

This left organizations in a bit of a quandary. Buy all-flash arrays now that offered the performance for their applications that needed it or wait until the data management on them were sufficiently mature?

X-IO’s iglu and ISE product lines address these concerns (which I why refer to their design as “thoughtful.”) Organizations may start with ISE 800 Series G3 All-Flash Array data storage system which offers the performance that many organizations initially want and need when deploying flash. It is with the data management features (LUN management, Web GUI, etc.) that organizations need to do base line management of the array. However it does not provide the fuller suite of features that organizations may need an array to offer before they deploy it more widely in their data center.

It is when organizations are ready to scale and use flash beyond just a point solution in their data center that the iglu blaze fx comes into play. The iglu introduces these more robust data management services that many organizations often need an array to offer to justify deploying it more broadly in their data center.

The decoupling of performance and data management services such as X-IO has done with its ISE Data Storage Systems and iglu Enterprise Storage Systems reflects a very thoughtful way for organizations to introduce flash into their environment as well as a means for X-IO to independently innovate and deliver on both data management and performance features without organizations having to unnecessarily pay for features they do not need.

The recent announcements about its flash optimized controllers and stretch clustering on its iglu blaze fx illustrate this mindset perfectly. Organizations that need raw performance can get that baseline functionality that flash offers by continuing to deploy the ISE 800 data storage system. But for those who are ready to use flash more widely in their data center, need more functionality than simple performance to achieve that goal and are ready to make the investment without sacrificing the investment in flash that they have already made, X-IO offers such a solution. That is what I call thoughtful.




It was All About the Petabytes at the NAB Show

A little over a decade ago when I told people that I was managing three (3) storage arrays with eleven (11) TBs of storage under management, people looked at me with a mixture of shock and awe. Fast forward to 2015 and last week’s NAB conference in Las Vegas, NV, and it was hard to find many storage vendors who even wanted to have a conversation with a customer unless it had at least a petabyte of data under management.

The NAB show is the media and entertainment industry’s largest event by far with an estimated 100,000+ people in attendance. (Yes, you read that correctly – one hundred thousand plus people.) Filling up every exhibit hall in the Las Vegas Conference Center with over 1 million square feet of exhibit space, vendors of every size, shape, type and flavor had their wares on display at the event that included everything from cameras to drones to high tech storage arrays.

As DCIG primarily focuses on data protection and data storage, I spent most of my time in the lower level of the south hall of the Las Vegas Convention Center catching up with the storage vendors who had exhibits. Some of the more interesting technologies on display included:

  • AMP Inc. (Accelerated Memory Production, Inc.) was promoting 4, 8 and even 16 TB SSD drives that would be available in 2.5” form factors in June. While I was looking at AMP Inc’s literature, another individual came up and engaged the AMP sales folks at the booth and expressed how much he was looking forward to the availability of these high capacity SSDs. Apparently, he is a camera man and needs SSDs with sufficient capacity to store all of the video he is recording in a single day without stopping. These new SSDs would apparently address his capacity concerns.

In my ensuing conversation that I had with Dan Stirling, AMP’s Director of Product Development and who was at the booth, he explained that AMP’s SSDs are primarily intended for industrial use applications. One use case is for the military where SSDs are deployed into environments that have large amounts of vibration and/or need to be removed and inserted frequently. These use cases and high capacity might also explain why no one at the AMP booth was willing to provide a quote on what these individual cartridges cost though, offhand, I guess $1000/TB is probably a good starting point for anyone looking at any of its SSDs.

  • The FlacheStreams FlacheSAN2 storage array at EchoStreams booth caught my attention if for no other reason than it had 48 2.5” SATA flash drives in a 2Us of rack space. If I read its specs correctly, the FlacheSAN2 in this configuration can sustain up to 250Gb/sec throughput and support up to 2 million IOPS.

Yet what more particularly caught my attention was that the vast majority of vendors who were exhibiting at NAB and with whom I spoke basically came right out and said that unless you have 1PB or more of data currently under management, they were not sure they had a product that would meet your needs. Granted, this was the NAB show and many organizations in the media and entertainment market need large amounts of storage capacity. But I do not recall vendors ever previously drawing such a clear line in the sand saying that in order to talk with them and for them to be of value to you, you really need to a petabyte of data under management or be well on your way to doing so.

Yet when I compare what these vendors are saying to what DCIG is finding in its own results from its various storage Buyer’s Guides, a petabyte of data is not even all that impressive anymore. In reviewing the results of the recently published DCIG 2015-16 Enterprise Midrange Array Buyer’s Guide, 33 of the 50+ models evaluated scaled to more than 2 PBs suggesting to me that the dividing line between large and midsize enterprise is already 2PBs, not the 1PB threshold that these vendors are using.

The NAB show is an event unlike almost any other in that you get a worm’s eye view of many of the technologies that are used in the production of the movies and TV shows that we all routinely watch. However the NAB show also provides a glimpse of where storage technology is going in the not too distant future as media and entertainment tend to push storage technologies in ways and directions that businesses typically do not. In that vein, it is clear that petabytes have officially replaced terabytes as the new benchmarks by which data is measured and that flash drives will become ever more prevalent with larger capacities sooner rather than later.




The Performance of a $500K Hybrid Storage Array Goes Toe-to-Toe with Million Dollar All-Flash and High End Storage Arrays

On March 17, 2015, the Storage Performance Council (SPC) updated its “Top Ten” list of SPC-2 results that includes performance metrics going back almost three (3) years to May 2012. Noteworthy in these updated results is that the three storage arrays ranked at the top are, in order, a high end mainframe-centric, monolithic storage array (the HP XP7, OEMed from Hitachi), an all-flash storage array (from startup Kaminario, the K2 box) and a hybrid storage array (Oracle ZFS Storage ZS4-4 Appliance). Making these performance results particularly interesting is that the hybrid storage array, the Oracle ZFS Storage ZS4-4 Appliance, can essentially go toe-to-toe from a performance perspective with both the million dollar HP XP7 and Kaminario K2 arrays and do so at approximately half of their cost.

Right now there is a great deal of debate in the storage industry about which of these three types of arrays – all-flash, high end or hybrid – can provide the highest levels of performance. In recent years, all-flash and high end storage arrays have generally gone neck-and-neck though all-flash arrays are generally now seen as taking the lead and pulling away.

However, when price becomes a factor (and when isn’t price a factor?) such that enterprises have to look at price and performance, suddenly hybrid storage arrays surface as very attractive alternatives for many enterprises. Granted, hybrid storage arrays may not provide all of the performance of either all-flash or high end arrays, but they can certainly deliver superior performance at a much lower cost.

This is what makes the recently updated Top Ten results on the SPC website so interesting. While the breadth of arrays covered in the published SPC results by no means cover every storage array on the market, they do provide enterprises with some valuable insight into:

  • How well hybrid storage arrays can potentially perform
  • How comparable their storage capacity is to high-end and all-flash arrays
  • How much more economical hybrid storage arrays are

In looking at these three arrays that currently sit atop the SPC-2 Top Ten list and how they were configured for this test, they were comparable in one of the ways that enterprises examine when making a buying decision. For instance, all three had comparable amounts of raw capacity.

Raw Capacity

High-End HP XP7                                                                         230TB
All-Flash Kaminario K2                                                              179TB
Hybrid  Oracle ZFS Storage ZS4-4 Appliance                    175TB

Despite using comparable amounts of raw capacity for testing purposes, they got to these raw capacity totals using decidedly different media. The high end, mainframe-centric HP XP7 used 768 300GB 15K SAS HDDs to get to its 230TB total while the all-flash Kaminario K2 used 224 solid state drives (SSDs) to get to its 179TB total. The Oracle ZS4-4 stood out from these other two storage arrays in two ways. First, it used 576 300GB 10K SAS HDDs. Second, its storage media costs were a fraction of the other two. Comparing strictly list prices, its media costs were only about 16% of the cost of the HP XP7 and 27% of the cost of the Kaminario K2.

These arrays also differed in terms of how many and what types of storage networking ports they each used. Both the HP XP7 and the Kaminario K2 used a total of 64 and 56 8Gb FC ports respectively for connectivity between the servers and their storage arrays. The Oracle ZS4-4 only needed 16 ports for connectivity though it used Infiniband for server-storage connectivity as opposed to 8Gb FC. The HP XP7 and Oracle ZS4-4 also used cache (512GB and ~3TB respectively) while the Kaminario K2 used no cache at all. It instead used a total of 224 solid state drives (SSDs) packaged in 28 flash nodes (8-800GB SSDs in each flash node.)

This is not meant to disparage the configuration or architecture of any of these three different storage arrays as each one uses proven technologies in the design of their arrays. Yet what is notable is the end results when these three arrays in these configurations are subjected to the same SPC2 performance benchmarking tests.

While the HP XP7 and Kaminario K2 came out on top from an overall performance perspective, it is interesting to note how well the Oracle ZS4-4 performs and what its price/performance ratio is when compared to the high end HP XP7 and the all-flash Kaminario K2. It provides 75% to over 90% of the performance of these other arrays at a cost per MB that is up to 46% less.

SPC-2 Top Ten ResultsSource: “Top Ten” SPC-2 Results, https://www.storageperformance.org/results/benchmark_results_spc2_top-ten

It is easy for enterprises to become enamored with all-flash arrays or remain transfixed on high-end arrays because of their proven and perceived performance characteristics and benefits. But these recent SPC-2 performance benchmarks illustrate that hybrid storage arrays such as the Oracle ZFS Storage ZS4-4 Appliance can deliver levels of performance that are comparable to million-dollar all-flash and high-end arrays at half of their cost which are numbers that any enterprise can take to the bank.




DCIG Announces Calendar of Planned Buyer’s Guide Releases in the First Half of 2015

At the beginning of 2014, I started the year with the theme: “it’s an exciting time to be part of the DCIG team“. This was due to the explosive growth we saw in website visits and popularity of our Buyer’s Guides. That hasn’t changed. DCIG Buyer’s Guides continue to grow in popularity, but what’s even more exciting is the diversity of our new products and services. This year’s theme is diversity: a range of different things. DCIG is expanding…again…in different directions.

In the past year, we have added a number of offerings to our repertoire of products and services.   In addition to producing our popular Buyer’s Guides and well known blogs, we now offer Competitive Research Services, Executive Interviews, Executive White papers, Lead Generation, Special Reports and Webinars. Even more unique, DCIG now offers an RFP/RFI Analysis Software Suite. This suite gives anyone (vendor, end-user or technology reseller) the ability to license the same software that DCIG uses internally to develop its Buyer’s Guide. In this way, you may use the software to do your internal technology assessments with your own scores and rankings so that the results align more closely with your specific business needs.

While we diversify our portfolio, it’s important to note that we also increased our Buyer’s Guide publication output by nearly 40% to thirteen (13) over our 2013 publications. We also contracted for over 30 Competitive Advantage reports in 2014.   This success is largely due the well-planned timeline, more clearly defined processes, and the addition of new analysts. The team is busy and here is a sneak peek at the Buyer’s Guides that they are currently working on during the first half of 2015 (in order of target release date):

Hybrid Storage Array: Hybrid Storage Array is a physical storage appliance that dynamically places data in a storage pool that combines flash memory and HDD storage (and in some cases NVRAM and/or DRAM) resources by intelligently caching data and metadata and/or by automatically moving data from one performance tier to another. The design goal of a hybrid storage array is to typically provide sub-2-millisecond response times associated with flash memory storage arrays with capacity and cost similar to HDD-based arrays.

SDS Server SAN: A new Buyer’s Guide for DCIG, the SDS Server SAN is a collection of servers combining compute, memory and internal DAS storage, which enables organizations to remove the need to for external storage in a virtualized environment. The SDS Server SAN software provides the glue between the compute and storage portions of the environment allowing for clustering of not only the virtual host but the underlying file system as well. SDS Server SAN’s typically bundle compute, storage and hypervisors and employ the usage of SSD as a tier for storage caching; SAS and/or SATA HDDs for data storage; and, support of one or more hypervisors.

Hybrid Cloud Backup Appliance: A Hybrid Cloud Backup Appliance is a physical appliance that comes prepackaged with server, storage and backup software. What makes this Buyer’s Guide stand apart from the Integrated Backup Appliances is that the Hybrid Cloud Backup Appliance must support backup both locally and to cloud providers. In this new Buyer’s Guide for DCIG, DCIG evaluates which cloud provider or providers that the appliance natively supports, the options it offers to backup to the cloud and even what options are available to recover data and/or applications with a cloud provider.

Private Cloud Storage Array: Private Cloud Storage Array is a physical storage appliance located behind an organization’s firewall that enables the delivery of storage as a service to end users within an enterprise. Private cloud storage brings the benefits of public cloud storage to the enterprise—rapid provisioning/de-provisioning on storage resources through self-service tools and automated management, scalability, and REST API support for cloud-native apps—while still meeting corporate data protection, security and compliance requirements

Flash Memory Storage Array: The Flash Memory Buyer’s Guide is a refresh from 2014. The flash array is a solid state storage disk system that contains multiple flash memory drives instead of hard disk drives.

Unified Communications: Another new guide for DCIG, Unified communications (UC) is any system that integrates real-time and non-real-time enterprise communication services such as voice, messaging, instant messaging, presence, audio and video conferencing and mobility features. The purpose of UC is to provide a consistent user-interface and experience across multiple devices and media-types.

Watch the latter half of the year as DCIG plans to refresh Buyer’s Guides on the following topics:

  • Big Data Tape Library
  • Deduplicating Backup Appliance
  • High End Storage Array
  • Integrated Backup Appliance
  • Midrange Unified Storage
  • SDS Storage Virtualization
  • Virtual Server Backup Software

We also have other topics that we are evaluating as the basis for new Buyer’s Guides so look for announcements on their availability in the latter half of this year.




DCIG Announces Calendar of Planned Buyer’s Guide Releases in the First Half of 2015

At the beginning of 2014, I started the year with the theme: “it’s an exciting time to be part of the DCIG team“. This was due to the explosive growth we saw in website visits and popularity of our Buyer’s Guides. That hasn’t changed. DCIG Buyer’s Guides continue to grow in popularity, but what’s even more exciting is the diversity of our new products and services. This year’s theme is diversity: a range of different things. DCIG is expanding…again…in different directions.

In the past year, we have added a number of offerings to our repertoire of products and services.   In addition to producing our popular Buyer’s Guides and well known blogs, we now offer Competitive Research Services, Executive Interviews, Executive White papers, Lead Generation, Special Reports and Webinars. Even more unique, DCIG now offers an RFP/RFI Analysis Software Suite. This suite gives anyone (vendor, end-user or technology reseller) the ability to license the same software that DCIG uses internally to develop its Buyer’s Guide. In this way, you may use the software to do your internal technology assessments with your own scores and rankings so that the results align more closely with your specific business needs.

While we diversify our portfolio, it’s important to note that we also increased our Buyer’s Guide publication output by nearly 40% to thirteen (13) over our 2013 publications. We also contracted for over 30 Competitive Advantage reports in 2014.   This success is largely due the well-planned timeline, more clearly defined processes, and the addition of new analysts. The team is busy and here is a sneak peek at the Buyer’s Guides that they are currently working on during the first half of 2015 (in order of target release date):

Hybrid Storage Array: Hybrid Storage Array is a physical storage appliance that dynamically places data in a storage pool that combines flash memory and HDD storage (and in some cases NVRAM and/or DRAM) resources by intelligently caching data and metadata and/or by automatically moving data from one performance tier to another. The design goal of a hybrid storage array is to typically provide sub-2-millisecond response times associated with flash memory storage arrays with capacity and cost similar to HDD-based arrays.

SDS Server SAN: A new Buyer’s Guide for DCIG, the SDS Server SAN is a collection of servers combining compute, memory and internal DAS storage, which enables organizations to remove the need to for external storage in a virtualized environment. The SDS Server SAN software provides the glue between the compute and storage portions of the environment allowing for clustering of not only the virtual host but the underlying file system as well. SDS Server SAN’s typically bundle compute, storage and hypervisors and employ the usage of SSD as a tier for storage caching; SAS and/or SATA HDDs for data storage; and, support of one or more hypervisors.

Hybrid Cloud Backup Appliance: A Hybrid Cloud Backup Appliance is a physical appliance that comes prepackaged with server, storage and backup software. What makes this Buyer’s Guide stand apart from the Integrated Backup Appliances is that the Hybrid Cloud Backup Appliance must support backup both locally and to cloud providers. In this new Buyer’s Guide for DCIG, DCIG evaluates which cloud provider or providers that the appliance natively supports, the options it offers to backup to the cloud and even what options are available to recover data and/or applications with a cloud provider.

Private Cloud Storage Array: Private Cloud Storage Array is a physical storage appliance located behind an organization’s firewall that enables the delivery of storage as a service to end users within an enterprise. Private cloud storage brings the benefits of public cloud storage to the enterprise—rapid provisioning/de-provisioning on storage resources through self-service tools and automated management, scalability, and REST API support for cloud-native apps—while still meeting corporate data protection, security and compliance requirements

Flash Memory Storage Array: The Flash Memory Buyer’s Guide is a refresh from 2014. The flash array is a solid state storage disk system that contains multiple flash memory drives instead of hard disk drives.

Unified Communications: Another new guide for DCIG, Unified communications (UC) is any system that integrates real-time and non-real-time enterprise communication services such as voice, messaging, instant messaging, presence, audio and video conferencing and mobility features. The purpose of UC is to provide a consistent user-interface and experience across multiple devices and media-types.

Watch the latter half of the year as DCIG plans to refresh Buyer’s Guides on the following topics:

  • Big Data Tape Library
  • Deduplicating Backup Appliance
  • High End Storage Array
  • Integrated Backup Appliance
  • Midrange Unified Storage
  • SDS Storage Virtualization
  • Virtual Server Backup Software

We also have other topics that we are evaluating as the basis for new Buyer’s Guides so look for announcements on their availability in the latter half of this year.




The Persisting Enterprise Reticence to Adopt Flash

Flash is by all estimates the future of enterprise production storage with most enterprises anticipating a day in the not too distant future where they will use flash storage arrays (all-flash or hybrid) much more broadly within their data center. Yet despite flash’s many benefits (higher levels of performance, smaller data center footprint and reduced energy consumption among others,) many enterprises still only use flash in a limited capacity if they use it at all. Today I take a look at some of the factors that still contribute to an enterprise reticence to adopt flash more broadly.

The enterprise reticence to adopt flash in any form stems from a number of a factors that include:

  • Market confusion. The DCIG 2014-14 Flash Memory Storage Array Buyer’s Guide identified more than 40 different all-flash array models from 20 different storage providers while the DCIG 2014 Hybrid Storage Array Buyer’s Guide identified another 40-plus models from 17 different providers. Having all of these different models available from which to choose make it difficult for enterprises to identify the most appropriate one for them.
  • Different types of flash. All-flash and hybrid storage arrays make one or more types flash memory available which may include MLC (multi-level cell), cMLC (consumer MLC), eMLC (enterprise MLC) and SLC (single level cell). Each of these flash memory types have characteristics that influence their cost, endurance, power consumption, total capacity and read and write performance. This requires that enterprises align their specific needs with the capabilities of these different flash media types.
  • Multiple deployment options. Flash memory may be placed on the server, in the network and on the storage array. Again, enterprises need to make a determination at what level in their infrastructure that they want to deploy flash.
  • No clear market leader. Enterprises tend to prefer to use storage technologies from clear market leaders. Yet as it relates to all-flash and hybrid storage arrays, a definitive market leader has yet to emerge. Many storage arrays models shipping today are either from a start-up or, if the model is available from an established storage provider, it is a net new model in their storage lineup that they have internally developed or is available as a result of acquiring a flash memory start-up such as Cisco and EMC have done.
  • Immaturity of data management services. Enterprises wants a high degree of certainty when it comes to the stability of the data on their production storage arrays. Since many all-flash and hybrid storage array models are either available from start-ups or are net new models from established providers, the maturity of the data management software and services on these arrays is often a question mark. These may offer few proof points and little is known about how well the data management services on these arrays actually works.
  • Managing flash in existing HDD-based arrays. To address these enterprise concerns about the immaturity of data management services, vendors are putting flash into existing HDD-based arrays. This approach does address concerns about the maturity of the array’s data management services. However these arrays were built for HDDs, not flash, so they may not include the functionality needed to manage flash or fully deliver on the performance improvements that flash has to offer.
  • Flash’s cost. Perhaps the biggest reason enterprises have put off adopting flash in the near-term is its relatively high upfront cost when compared to HDDs. While the price per raw GB for flash has come down substantially in recent years, its cost can still be a factor of anywhere from 5 to 20 x more than HDDs though technologies such as compression and deduplication are helping to bring the cost of flash down to make it feasible for more enterprises to adopt.



Oracle Brings out the Big Guns, Rolls out the FS1 Flash Storage System

Dedicating a single flash-based storage array to improving the performance of a single application may be appropriate for siloed or small SAN environments. However this is NOT an architecture that enterprises want to leverage when hosting multiple applications in larger SAN environments, especially if the flash-based array has only a few or unproven data management services behind it. The new Oracle FS1 Series Flash Storage System addresses these concerns by providing enterprises the levels of performance and the mature and robust data management services that they need to move flash-based arrays from the fringes of their SAN environments into their core.

Throwing flash memory at existing storage performance problems has been the de facto approach of most hybrid and flash memory storage arrays released to date. While these flash-based arrays deliver performance improvements of 3x to as much as 20x or more over traditional hard disk drive (HDD) based arrays, they are frequently deployed as a single array dedicated to improving the performance of a single application.

This approach breaks down in enterprise environments that want to attach multiple (and ideally all) applications to a single flash-based array. While performance improvements will likely still occur in this scenario, many flash-based arrays often lack any intelligence to prioritize I/O originating from different applications.

This results in I/O traffic from higher priority applications being given the same priority as lower priority applications based on the assumption that flash-based arrays are “so fast” that they will be able to service all I/O from all applications equally well. Meanwhile, I/O from mission-critical applications wait as I/O from lower priority applications get served.

The drawbacks with this approach are three-fold:

  • Enterprises want a guarantee that their most mission critical applications such as Oracle E-Business Suite will get the performance that they need when they need it over other, lower tier applications. In today’s environments, no such guarantees exist.
  • Should an application’s performance requirements change over time, today’s flash-based storage arrays have no way to natively detect these changes.
  • Business owners of lower-tier applications will internally campaign to connect their applications to these flash-based arrays as they will, by default, get the same performance as higher tier applications. This further impacts the ability of mission-critical applications to get their I/Os served in a timely manner.

Even should these arrays deliver the performance that all of these applications needs, the data management services they offer are either, at best, immature or, at worst, incomplete and insufficient to meet enterprise demands. This is why today’s flash-based arrays fall short and what sets the Oracle FS1 Series Flash Storage System apart.

The Oracle FS1 was specifically architected for flash media as its primary function with HDD support a secondary focus. Further distinguishing it from other all-flash arrays, the FS1 offers up to two tiers of flash with an optional, additional two tiers of disk to provide a four-tier storage architecture with data intelligently and automatically moved between tiers.

The Oracle FS1 comes equipped with the specific technologies that today’s enterprises need to justify deploying a flash storage array into the heart of their SAN environment. It can simultaneously and successfully handle multiple different application workloads with the high levels of predictable performance that meets their specific needs.

On the hardware side, it delivers the high end specifications that enterprises expect. It is architected to support as many as sixteen (16) nodes in a single logical system, enterprises may start with an FS1 configuration as small as two (2) nodes and then scale it to host petabytes of flash capacity in a single logical configuration. This is more than twice the size of EMC’s XtremeIO (6 nodes) or an HP 3PAR StoreServ 7400 (8 nodes). In the 16 node configuration, internal Oracle tests have already shown the FS1 capable of supporting up to 80 GBps of throughput and 2 million IOPs.

It is on the software side with its native data management services that puts the FS1 in a class by itself. Most flash-based storage arrays have either minimal data management services or, if they do offer them, the data management services are, at best, immature. The Oracle FS1 provides the mature, full suite of data management services that enterprises want and need to justify deploying a flash-based solution into their SAN environments.

Further, Oracle makes it easy and practical for enterprises to take advantage of this extensive set of data management services as they are included with every Oracle FS1 as part of the base system. In this way, any enterprise that deploys an Oracle FS1 has immediate access to its rich set of software features.

Consider these highlights:

  • Storage profiles associated with each application. The key to prioritizing application I/O and putting the right data on the right disk is to first establish the application priority and then associate it with the right tier or tiers of disk. To deliver on this requirement, the Oracle FS1 offers pre-defined, pre-tested and pre-tuned storage profiles that are optimized for specific applications.

Using these profiles, enterprises may, with a single click, associate each application with a specific storage profile that is optimized for that application’s specific capacity, performance and cost requirements. For example, demanding Oracle Database applications may be provisioned with “Premium Priority” high performance storage profiles that consist of tiers of flash disk. This priority level ensures that mission-critical applications receive the low latency service they require. Conversely, lower tier, less demanding applications may be associated with medium or low priority storage profiles that provision tiers of performance and capacity-oriented HDDs.

  • Application I/O prioritization. Associating applications with storage profiles eliminates the need for the Oracle FS1 to rely on the traditional “cross your fingers and hope for the best” means of application I/O prioritization. Knowing the priority of applications enables the FS1 to receive and prioritize I/Os according to the application sending them.

As it simultaneously receives I/Os from multiple different applications, it recognizes which I/Os are associated with high priority applications and services them first. This “priority in, priority out” option eliminates the risk and uncertainty associated with the “first in, first out” methodology predominantly found on most flash-based arrays today.

  • Adds business value to application I/O management. The prevalent I/O queue management technique used in flash and HDD storage systems is “first in, first out” just as it was with the first hard disk drive – the IBM 305 RAMAC – in 1958.

The world has changed a bit since then. The Oracle FS1 recognizes that different applications have different value to the enterprise and the Oracle FS1’s QoS Plus takes into this into account as it prioritizes I/O. As the FS1 receives I/Os from multiple different applications, it recognizes which I/Os are associated with “Premium Priority” (high business value) applications and services them first. This “priority in, priority out” option eliminates the risk and uncertainty associated with the “first in, first out” I/O queue management.

The Oracle FS1 QoS Plus delivers further business value by placing data across up to four different storage layers (performance and capacity flash media along with performance and capacity HDDs.). QoS Plus collects detailed information on the applications’ storage usage profile, evaluates data for movement to different storage tiers, then combines that with auto-tiering to automatically migrate data to the most cost-effective media (flash or disk) from a $/IOP and $/GB standpoint based on the application data’s usage profile AND the value of that data to the business.

  • Capacity optimization. To dynamically optimize data placement on available storage capacity, the Oracle FS1 stores data in 640K chunks. Storing data in these size chunks, it optimally uses the available flash and HDD storage capacity in the FS1 without creating too much management overhead on the system as can happen using smaller, 4K chunks. This also minimizes the waste that can occur at the other extreme as some storage arrays store and move data in chunks as large as 1GB (1600x larger than FS1.)

The FS1 then tracks the performance of individual 640K chunks over time using workload-driven heat maps. If chunks that reside on flash are infrequently or not accessed, they may get moved down to lower tiers of flash or disk; conversely, chunks that reside on HDDs may become more active over time so they may get moved to a higher tier of disk or even flash.

  • Isolate data in containers. The Oracle FS1’s Storage Domains software enables the creation of multiple, virtual storage systems within a single Oracle FS1, a feature not readily available with flash storage systems. Each storage domain is a “data container” which isolates data from other storage domains.

The Storage Domains multiple unique environments with individual custom-tailored QoS Plus settings can reside on a single physical FS1, reducing power, cooling, and management administration expense. This multi-tenancy capability is ideal for private or public cloud deployments, regulatory compliance requirements, or chargeback models.

  • Optimized for Oracle Database environments. The Oracle FS1 Series supports all major operating systems, hypervisors and applications. However enterprises running Oracle Database in their environment will experience benefits that no other vendor’s flash- or HDD-based array can offer.

By supporting Oracle Database’s Advanced Data Optimization (ADO) and Hybrid Columnar Compression (HCC), enterprises achieve levels of performance and capacity optimization for Oracle Database that other vendors’ flash-based arrays cannot provide because they are not co-engineered for deep levels of integration with Oracle Database.

The Oracle FS1 Series Flash Storage System breaks new ground in flash-based array SAN battleground by delivering more than high levels of performance which is often where other flash-based storage arrays start and stop. The Oracle FS1 stands apart from its competitors in this space by providing a highly available and scalable architecture backed by mature and proven suite of data management services that are part of its base system, not separately licensed options. With Oracle FS1, enterprises can finally move ahead with their plans to bring flash storage arrays to run multiple applications and workloads at the core of their SAN environments, not just as single-application point products.




What Will Be Hot in Flash in the Years to Come

A couple of weeks ago I attended the Flash Memory Summit in Santa Clara, CA, where I had the opportunity to talk to a number of providers, fellow analysts and developers in attendance about the topic of flash memory. The focus of many of these conversations was less about what flash means right now as its performance ramifications are already pretty well understood by the enterprise. Rather many are already looking ahead to take further advantage of flash’s particular idiosyncrasies and, in so doing, give us some good insight into what will be hot in flash in the years to come.

Everyone in the storage industry knows that flash is not spinning media. If anything, the fact that everyone is largely forced to treat flash as storage is arguably one of flash’s big drawbacks to date. Exacerbating the issue, storage vendors are still going through a learning curve of their own. When they think of storage, many are still predisposed to approach it and think of it in an “HDD First” mentality. While that is only to be expected after years of designing storage systems for HDDs, a new “Flash First” mentality is now required and different vendors are at various stages internally at making this shift in thinking.

This is what made the Flash Memory Summit particularly intellectually stimulating to attend. A fair number of the vendors at the conference approach and think of flash in this context of “Flash First.” As such, they are pushing the boundaries of flash and while their ideas and products are, in most cases, probably not ready for enterprise adoption or use today, they do provide some good insight into what flash is poised to deliver in just a few years.

1.  No Read Penalty. Everyone always talks about the write penalty associated with flash and how their systems manage that. But everyone has been so obsessed with overcoming and dealing with this issue is that they have overlooked a big opportunity with flash: there is no read penalty.

Once data is written to flash, whether you read a piece of data once, a hundred times, a million times or more, there is no penalty (as in there is no penalty to be paid in the wearing out of the flash.) This is a significant divergence from HDDs which are mechanical devices and have fairly defined wear out periods. Whether you are reading from or writing to HDDs, if you get 5 years out of an HDD, that is about it.

Acknowledging and taking advantage of this “No Read Penalty” property of flash is going to force a major rethinking of flash’s role for use as a preferred media in archive. Unlike HDDs, flash has the potential to last for many more years (potentially 10 or more,) provide 5-10x improvements in read performance over HDDs, use less power, produce less heat and consume far less rack space due to its density.

Already I saw a 64 TB (yes, TB) flash memory card being demo’d at the Flash Memory Summit by NxGn Data that fits into a PCI slot. This card is strictly intended for Read intensive environments where one write or maybe at most a few writes occur to the card. After that, the rest of the I/Os are read traffic with the first version of its card able to achieve read I/Os in the 250,000 – 300,000 IOPS range. Using this type of technology. I can already think of a number of Big Data applications that can benefit from it. Using it, organizations can put data that never changes (videos, photos, images, etc.) on these cards to get unparalleled read performance at a fraction of the capital cost of today’s flash solutions.

2. More Application Integration is Coming. Another major challenge with flash is that applications and operating systems have yet to adapt to take advantage of the performance that flash currently offers. Until now, applications and OSes had to be tuned to account for IO queue depths and IO wait times due to slow storage response times.

Those days are over. Now it is the storage that is waiting on the applications and OSes to send it more IOs and more data to process. This sudden availability of storage performance is prompting the redesigns applications and OSes so that they can begin to take advantage of storage performance in ways they never could before.

To do so, integration has to occur between the applications and the OSes and the underlying storage to facilitate faster data transfers and increased throughput. Already we see early examples of this from Oracle between Oracle Database and its ZFS storage solutions and IBM DB2 and its storage systems. Look for much more of this type of integration in the months and years to come and that by 2016, most leading storage arrays will offer some type of integration with leading applications and operating systems.

3. The Demise (or End?) of Storage as We Know It. One major theme that came out of the Flash Memory Summit was the current inefficiency of needing to translate data into a format that storage understands and then back into a format that flash understands. The thought was: why not just create a bus from flash memory directly into the CPU to avoid all of these theoretically unneeded translations of data back and forth between these different formats?

As the situation stands now, the translation of data between the CPU and underlying flash-based storage occurs very quickly today – so quickly that most applications and OSes cannot detect the delay. But as applications and OSes catch up and again get out in front of today’s flash-based storage solutions, look for a new generation of storage to emerge that may not even appear as storage at all but as ever-present memory cache. Forerunners of solutions like this already ship from Fusion-io (now a part of SanDisk) and I would expect more to emerge in the years to come as the benefits of storing data in this format become more apparent.




The Challenges of Delivering Inline Deduplication on a High Performance Production Storage Array

The use of data reduction technologies such as compression and deduplication to reduce storage costs are nothing new. Tape drives have used compression for decades to increase backup data densities on tape while many modern deduplicating backup appliances use compression and deduplication to also reduce backup data stores. Even a select number of existing HDD-based storage arrays use data compression and deduplication to minimize data stores for large amounts of file data stored in archives or on networked attached file servers.

The challenges of using these two technologies change when they are implemented in high performance environments. The more predictable data access patterns with lots of redundant data that exist in archive, backup, and, to some extent, file serving environments are replaced in high performance environments with applications that potentially have highly random data access patterns where data does not deduplicate as well. Capacity reductions of production data are not as significant (maybe in the 2-3x range) as in backup which can achieve deduplication ratios of up to 8-20X or even higher.

Aggravating the situation, there is little to no tolerance for performance interruptions in the processing of production data – raw or deduplicated. While organizations may tolerate the occasional slow periods of deduplication performance for archive, backup and file servers data stores, consistently high levels of application performance with no interruptions are the expectations here.

Yet when it comes to deduplicating data, there is a large potential for a performance hit. In high performance production environments with high data change rates and few or no periods of application inactivity, all deduplication must be done inline. This requires the analysis of incoming data by breaking packets of data apart into smaller chunks, creating a hash and comparing that hash to existing hashes in the deduplication metadata database to determine if that chunk of data is unique or a duplicate.

If the array determines a chunk of data is a duplicate, there is also a very small chance that a hash collision could occur. Should the all-flash array fail to detect and appropriately handle this collision, data may be compromised.

These expectations for high levels of data integrity and performance requires large amount of cache or DRAM to host the deduplication metadata. Yet all-flash storage arrays only contain fixed amounts of DRAM. This may limit the maximum amount of flash storage capacity on the array as it makes no sense for the array to offer flash storage capacity beyond the amount of data that it can effectively deduplicate.

These all-flash array capacity limits are reflected in the results of the most recent DCIG 2014-15 Flash Memory Storage Array Buyer’s Guide. Of the 36 all-flash array models evaluated, only 42 percent of them could scale to 100 TB or more of flash capacity. Of these models that could scale to more than 100 TB, they:

  • Did not support the use of data deduplication at the time the Guide was published
  • Did not publicly publish any performance data with deduplication turned “On” implying that they recommend turning deduplication “Off” when hosting performance sensitive applications
  • Use scale-out architectures with high node counts (up to 100) that are unlikely to be used in most production environments

The need to scale to 100 TBs or more of flash storage capacity is quickly becoming a priority. HP reports that already 25% of its HP 3PAR StoreServ 7450 all-flash arrays ship with 80TBs or more of raw capacity as its customers want to move more than just their high performance production data from HDDs to flash. They want to store all of their production data on flash. Further, turning deduplication off for any reason when hosting high performance application on these arrays is counter intuitive since these arrays are specifically designed and intended to host high performance applications. This is why, as organizations look to acquire all-flash storage arrays to host multiple applications in their environment, they need to look at how well they optimize both capacity and performance to keep their costs under control.




Today it is Really All About the Integrated Solution

As I attended sessions at Microsoft TechEd 2014 last week and talked with people in the exhibit hall a number of themes emerged including “mobile first, cloud first”, hybrid cloud, migration to the cloud, disaster recovery as a service, and flash memory storage as a game-changer in the data center. But as I reflect on the entire experience, a statement made John Loveall, Principal Program Manager for Microsoft Windows Server during one of his presentations sums up to overall message of the conference, “Today it is really all about the integrated solution.”

The rise of the pre-integrated appliance in enterprise IT has certainly not gone unnoticed by DCIG. Indeed, we have developed multiple buyer’s guides to help businesses understand the marketplace for these appliances and accelerate informed purchase decisions.

The new IT service imperative is time to deployment. Once a business case has been made for implementing a new service, every week that passes before the service is in production is viewed by the business as a missed revenue growth or cost savings opportunity—because that is what it is. The opportunity costs associated with IT staff researching, purchasing, integrating and testing all the components of a solution in many cases outweigh any potential cost savings.

An appliance-based approach to IT shrinks the time to deployment. The key elements of a rapidly deployable appliance-based solution include pre-configured hardware and software that has been pre-validated to work well together and then tested prior to being shipped to the customer. In many cases the appliance vendor also provides a simplified management tool that facilitates the rapid deployment and management of the service.

Some vendors in the TechEd exhibit hall that exemplify this appliance-based approach included DataOn, HVEconneXions, InMage, Nutanix and Violin Memory.

DataOn was previewing their next-generation Cluster-in-a-Box. Although the DataOn booth was showing their products pre-configured with Windows Server 2012 R2 and Storage Spaces, they also support other operating environments and are Nexenta certified. Nutanix takes a similar approach to deliver what they call a “radically simple converged infrastructure”.

I met the David Harmon, President of HVE ConneXions at the Huawei booth. HVE is using Huawei networking gear in combination with HVE’s own flash memory appliances to deliver VMware View-based virtual desktops to clients at a cost of around $200 per desktop. He told me of a pilot implementation where two HVE staff converted a 100 computer lab of Windows XP desktops to Windows 7 virtual desktops in just two days.

InMage Systems was showing their InMage 4000 all-in-one purpose-built backup and disaster recovery appliance that can also provide public and private cloud migration. I spoke with Joel Ferman, VP of Marketing, who told me that their technology is used by Cisco, HP and Sunguard AS; and that they had never lost a head-to-head proof of concept for either backup or disaster recovery. InMage claims their solution can be deployed in less than a day with no downtime. The appliance won the Windows IT Pro Best of TechEd 2014 award in the Backup & Recovery category.

Violin Memory was displaying their Windows Flash Array, an appliance that ships with Windows Storage Server 2012 R2 pre-installed. The benefits of this appliance-based approach was explained by Eric Herzog, Violin Memory’s CMO this way, “Customers do not need to buy Windows Storage Server, they do not need to buy blade servers, nor do they need to buy the RDMA 10-gig-embedded NICs. Those all come prepackaged in the array ready to go and we do Level 1 and Level 2 support on Windows Server 2012 R2.”

Today it is really all about the integrated solution. In many cases, the opportunity to speed the time to deployment is the deciding factor in selecting an appliance-based solution. In other cases, the availability of a pre-configured appliance puts sophisticated capabilities within reach of smaller IT departments composed primarily of IT generalists who lack the specialized technical skills required to assemble such solutions on their own. In either case, the ultimate benefit is that businesses gain the IT capabilities they need with a minimum investment of time.

This is the second in a series of blog entries based on my experience at Microsoft TechEd 2014. The first entry focused on how Microsoft’s inclusion of Storage Spaces software in Windows Server 2012 R2 paves the way for server SAN, and how Microsoft Azure Site Recovery and StorSimple promote hybrid cloud adoption.




Wikibon Might Be Right On in Its Forecast for Server SAN Growth After All

Toward the end of April Wikibon’s David Floyer posted an article on the topic of server SANs entitled “The Rise of Server SANs” which generated a fair amount of attention and was even the focus of a number of conversations that I had at this past week’s Symantec Vision 2014 conference in Las Vegas. However I have to admit, when I first glanced at some of the forecasts and charts that were included in that piece, I thought Wikibon was smoking pot and brushed it off. But after having had some lengthy conversations with attendees at Symantec Vision, I can certainly see why Wikibon made some of the claims that it did.

While I recommend you read the article at Wikibon’s site in its entirety, I’ll briefly summarize it. Floyer alleges that the combination of powerful servers with commodity hardware, server-side flash and storage software installed on the server (one possible iteration of software defined storage or SDS) will disrupt the current SAN and NAS attached storage array market of today. In its place two types of server SAN architectures will emerge: Hyperscale Server SAN Storage and Enteprise Server SAN storage. The collective impact of these two technologies over the next decade will be so dramatic that by 2024 they will generate up to 10x or more revenue than today’s storage arrays.

Assuming this forecast is true (and it is a pretty damn good reading of the tea leaves if it is,) the entire storage industry is headed for some pretty tumultuous times over the next decade. Based upon these predictions, today’s SAN and NAS attached systems will begin to feel a noticeable impact by as soon as 2017 with the sale of externally attached storage arrays falling off the cliff from 2019-2022 before finally bottoming out in the 2024-2027 time frame.

While these types of forecasts make for entertaining reading and good press, I initially dismissed this forecast as being too extreme to be realistic. But as I was at Symantec Vision this past week, I spent a great deal of time talking to the Storage Foundation team about server SANs and flash in general and its recently released (December 2013) SmartIO technology in particular which could contribute to Wikibon’s forecast for server SANs becoming a reality.

As I covered in a blog entry on SmartIO back in December 2013, SmartIO provides organizations with three distinct benefits:

  1. Organizations may do targeted deployments of in-server flash without exposing themselves to the risk of having a single copy of data on the server.
  2. Organizations get the performance benefits of in-server flash.
  3. Organizations may opt to use in-server flash on a larger scale by deploying it in conjunction with Tier 2 storage arrays.

Applying these three benefits to the concept of server SANs and their forecasted growth, we begin to see why technologies like SmartIO can make this a reality:

  1. Another significant leap forward in performance. Companies are currently deploying hybrid or all-flash arrays because they give them a 2-5x increase in performance over HDD-based arrays by reducing latency from 5-10 milliseconds to 1-2 milliseconds (or less.)  Using server SANs with server-side flash deployments, latencies are now measured in microsecond and even nanosecond response times.
  2. It’s much cheaper. SANs and NAS are very pricey but their deployments to data have been justified by lowering costs through shared storage, better utilization rates, and improved performance. Those benefits are almost completely eroded away by server SANs. Server side flash already provides better performance at a lower cost. While storage utilization rates may not (and I emphasize may not) be as good as in NAS and SAN environments, the savings realized through the deployment of server SANs may make this last point irrelevant (at least for many organizations.)
  3. It’s scales more simply and affordably. Need more performance, add another server with more flash drives or insert flash into existing servers. This is  simpler and more affordable to scale than adding an entire new storage array to the mix when an existing one runs out of capacity, performance or both.

As promising as server SAN technology looks, let’s not forget that this transition is not going to happen overnight. Here are a few key issues that proponents of server SAN architectures have yet to answer:

  1. Who you gonna call? Almost every organization large or small wants the assurance that their preferred storage vendor will fly in the black helicopters when applications or equipment break (and gear will break – count on it!) It is not clear yet if all server SAN providers are ready to step up to this challenge.
  2. Plug-n-play. Deploying server SANs that require any kind of touch on the server such as installing new software or hardware tends to be met with resistance by server admins. Further, not every server OS or server hardware can have just any software or hardware installed into it. SAN and NAS architectures do have the luxury of having an incumbent, in-place architecture where these types of battles (installs of HBAs, multi-pathing software, etc.) have already been largely fought and won. This makes the introduction of new all-flash or hybrid storage arrays into these environments today simpler and less intrusive and creates a barrier that server SANs need to overcome.
  3. Unproven. Save the example of Symantec’s Storage Foundation suite and its SmartIO, many competitors with server SAN solutions offer software that is still new and largely unproven. This alone will give any organization pause.

Having had some time to think and talk about this topic of Server SANs for the past week, it is clear that this storage architecture has a much brighter future than for which I initially gave it credit. That said, to say that it will crush SAN and NAS architectures over the next decade to the point where they will have, at best, a nascent percentage of the storage space assumes that everything breaks almost perfectly for server SAN providers. My gut feeling, having worked as a storage guy for 10 years and now having written about it for another 10, is that the only assumption you can make is that stuff will break and then you just hope that either it breaks in your favor or have a contingency plan in place to help ensure it does.




HP ProLiant BL660c VMware VMmark Benchmark Using HP 3PAR StoreServ 7450 All-flash Array Carries Real-World VM Density and Performance Implications

VMware® VMmark® has quickly become a performance benchmark to which many organizations turn to quantify how many virtual machines (VMs) they can realistically expect to host and then perform well on a cluster of  physical servers. Yet a published VMmark score for a specified hardware configuration may overstate or, conversely, fail to fully reflect the  particular solution’s VM consolidation and performance capabilities. The HP ProLiant BL660c published VMmark performance benchmarks using a backend HP 3PAR StoreServ 7450 all-flash array provide the relevant, real-world results that organizations need to achieve maximum VM density levels, maintain or even improve VM performance as they scale and control costs as they grow.

VMware VMmark Begins the Transformation of Virtualization Performance Measurement from an Art into a Science

Server virtualization is becoming, and largely has already become, the de facto way in which organizations introduce new applications into their IT infrastructure. Yet the techniques associated with forecasting the right number of VMs to host on each physical machine remains as much of an art as it is a science due to the variables involved (application type, server and storage hardware , etc.) that influence performance in a virtualized environment.

The VMware VMmark performance benchmark mitigates some of this guesswork associated with determining how many VMs a particular configuration can host. Introduced in 2007, the VMware VMmark virtualization platform scores performance in virtualized environments with mixed workloads by grouping VMs into tiles (the metric used by VMmark to benchmark virtualization performance,) measuring how quickly each tile completes specific tasks and then averaging each tile’s scores to provide an overall score.

The current VMmark 2.X increased the number of VMs in a tile from 6 to 8 and included additional tools for benchmarking multi-server virtualization environments. For example, VMmark 2.x measures the performance impacts of virtualization-specific workloads such as vMotion and storage vMotion on the underlying platform. The recent VMmark 2.5.x even provides specifications to measure power usage in virtualized environments.

These iterations of the VMware VMmark performance benchmark have collectively solved the following challenges for organizations by enabling them to:

  1. Objectively quantify how many VMs a physical machine can provide a sufficient, reliable amount of performance to host.
  2. View and compare the performance of different hardware platforms based upon published metrics
  3. Better forecast and budget for the appropriate amount of hardware to host their VMs

The Storage Variable

Despite the standardization that the VMmark performance benchmark has introduced into the previously subjective areas of measuring performance and power utilization in virtualized environments, variables still exist in how tested hardware solutions may be configured.

An allowance in the hardware configuration that the VMmark specification makes is permitting the use of various storage configurations when running these benchmarks. While this may seem like a nit, storage is often one of the leading contributors to performance bottleneck in many virtualized environments. As such, using a storage configuration that does not reflect a real world environment impacts how one should interpret published VMmark performance results. Two specific conditions to examine in these performance results to understand their relevance in the real world include:

  1. Storage configuration may not reflect real-world production conditions. Organizations often use external storage arrays to host production VMs. However many of these VMmark benchmark tests are conducted using either internal flash drives or external storage arrays with their flash or hard disk drives configured as RAID 0. While these storage configurations contribute to improved performance for the purposes of the VMmark performance benchmark, they do not represent what most organizations use in their production environments.
  2. Storage array performance capabilities not fully measured. The intent of the VMmark performance benchmark is to provide guidance as to the number of VMs with mixed workloads to which the server hardware can provide sufficient and reliable performance. While the storage array contributes to and influences performance, VMmark does not attempt to examine the specific performance capabilities of the underlying storage array. As such, it is both possible and probable that the full performance capabilities of the storage array are not fully tapped.

The Distinguishing Characteristics of the HP ProLiant VMmark Benchmarks

One of the notable exceptions to the many published VMmark benchmarks is the recent April 15, 2014, disclosure of the HP ProLiant BL660c Gen8 VMmark benchmarks. These HP ProLiant VMmark benchmarks distinguish themselves from other VMmark benchmark results in the following three ways:

  • HP 3PAR StoreServ 7450 utilized in the benchmarking. The HP 3PAR StoreServ 7450 is recognized by DCIG at a Top 3 enterprise all-flash storage array and is routinely used by organizations in their production environments. By using the HP 3PAR StoreServ 7450 as part of the HP ProLiant BL660c Gen8 VMmark benchmark, the published results provide a more realistic representation of the high levels of performance and VM density that an organization could expect of the HP ProLiant BL660c if deployed in their production environment since it uses a production storage array.
  • RAID 1 storage configuration used on the HP 3PAR StoreServ 7450. Almost all organizations without exception put the disk drives in their production storage array into some type of RAID configuration for data protection. In the HP VMmark benchmarks, a RAID 1 configuration was used for all of the tests as opposed to a RAID 0 configuration. This RAID 1 configuration more accurately represents the type of RAID configuration that would be found and used in a production storage array deployment.
  • More HP ProLiant BL660c Gen8 servers could be introduced, use the same HP 3PAR StoreServ 7450 and achieve comparable VM density and performance results. The published April 15, 2014, HP ProLiant BL660c VMmark benchmarks only include the performance results of two HP ProLiant hosts. These results suggest that an organization could potentially host up to 192 VMs on these two blade servers with each of these 192 VMs having access to very high levels of performance.

However there is nothing in these results to suggest that these two HP ProLiant servers were in any way close to maxing out the performance capabilities of the HP 3PAR 7450 array. This gives organizations the flexibility to connect more physical HP ProLiant BL660c servers to the HP 3PAR StoreServ 7450 array with the knowledge that they can achieve approximately the same levels of VM density and performance on these additional ProLiant servers as they did on their first two ProLiant servers without needing to acquire additional storage.

HP ProLiant BL660c Published VMmark Benchmarks Answer Call for Performance Tests that Have Real World Application

To say that data centers in general and virtualized data centers specifically need meaningful performance benchmarks in order to make better capacity planning and buying decisions is an understatement. In that sense, the VMware VMmark performance benchmarks provide an important step forward in providing organizations with the information they need to better assess and optimally utilize the hardware intended for hosting VMs in their environment.

However any performance benchmark and its application in production virtualized environments is only as useful as the underlying hardware configuration on which that benchmark is based.  The most recent HP ProLiant BL660c Gen8 VMmark benchmarks provide these sought after real-world performance metrics that enterprise virtualized environments want and need. While all organizations should always test new solutions in their environment before deploying them if at all possible, the use of the HP 3PAR StoreServ 7450 in these published VMmark benchmarks give organizations more reason than normal to be optimistic about the potential VM density and performance benefits that the 7450 can realistically deliver in their environment.

Bitnami