was successfully added to your cart.

SaaS Provider Pulls Back the Curtain on its Backup Experience with Cohesity; Interview with System Architect, Fidel Michieli, Part 3

Usually when I talk to backup and system administrators, they willingly talk about how great a product installation was. But it then becomes almost impossible to find anyone who wants to comment about what life is like after their backup appliance is installed. This blog entry represents a bit of anomaly in that someone willingly pulled back the curtain on what their experience was like after they had the appliance installed. In this third installment in my interview series with system architect, Fidel Michieli, describes how the implementation of Cohesity went in his environment and how Cohesity responded to issues that arose.

Jerome:  Once you had Cohesity deployed in your environment, can you provide some insights into how it operated and how upgrades went?

Fidel:  We have been through the upgrade process and the process of adding nodes twice. Those were the scary milestones that we did not test during the proof of concept (POC). Well, we did cover the upgrade process, but we did not cover adding nodes.

Jerome:  How did those upgrade go? Seamlessly?

Fidel:  The fact that our backup windows are small and we can run during the night essentially leaves all of our backup infrastructure idle during the day. If we take down one node at a time, we barely notice as we do not have anything running. But as software company, we expect there to be a few bumps along the way which we encountered.

Jerome:  Can you describe a bit about the “bumps” that you encountered?

Fidel:  We filled up the Cohesity cluster much faster than we expected which set its metadata sprawling. We went to 90-92 percent very quickly so we had to add in nodes in order to get the capacity back which was being taken up by its metadata.

Jerome:  Do you control how much metadata the Cohesity cluster creates?

Fidel:  The metadata size is associated with the amount of duplicated data it holds. As that grew, we started seeing some services restart and we got alerts of services restarting.

Jerome:  You corrected the out of capacity condition by adding more nodes?

Fidel:   Temporarily, yes.  Cohesity recognized we were not in a stable state and they did not want us to have a problem so they shipped us eight more nodes for us to create a new cluster.  [Editor’s Note:  Cohesity subsequently issued a new software release to store dedupe metadata more efficiently, which has since been implemented at this SaaS provider’s site.]

Jerome:  That means a lot that Cohesity stepped up to the plate to support its product.

Fidel:   It did. But while it was great that they shipped us the new cluster, I did not have any additional Ethernet ports to connect these new nodes as we did not have the additional port count in our infrastructure. To resolve this, Cohesity agreed to ship us the networking gear we needed. It talked to my network architect, found out what networking gear we liked, agreed to buy it and then shipped the gear to us overnight.

Further, my Cohesity system engineer, calls me every time I open a support ticket and shows up here. He replies and makes sure that my ticket moves through the support queue. He came down to install the original Cohesity cluster and the upgrades to the cluster, which we have been through twice already. The support experience has been fantastic and Cohesity has taken all of my requests into consideration as it has released software upgrades to its product, which is great.

Jerome:  Can you share one of your requests that Cohesity has implemented into its software?

Fidel:  We needed to have connectivity to Iron Mountain’s cloud. Cohesity got that certified with Iron Mountain so it works in a turnkey fashion. We also needed support for SQL Server which Cohesity into its road map at the time and which it recently delivered. We also needed Cohesity to certify support for Exchange 2016 so they expedited support for that so it is also now certified.

In part 1 of this interview series, Fidel shares the challenges that his company faced with its existing backup configuration as well as the struggles it encountered in identifying a backup solution that scaled to meet a dynamically changing and growing environment.

In part 2 of this interview series Fidel shares how he gained a comfort level with Cohesity prior to rolling it out enterprise-wide in his organization.

In part 4 of this interview series Fidel shares how Cohesity functions as both an integrated backup software appliance and a deduplicating target backup appliance in his company’s environment.

image_pdfimage_print
Jerome M. Wendt

About Jerome M. Wendt

President & Lead Analyst of DCIG, Inc. Jerome Wendt is the President and Lead Analyst of DCIG Inc., an independent storage analyst and consulting firm. Mr. Wendt founded the company in September 2006.

Leave a Reply