Life Left in “End of Life” XtremIO X1 Storage Systems

The XtremIO downsides do underscore the importance of excellent maintenance. Response to any first sign of failure must be addressed quickly to protect the node and its data.

The announcement that XtremIO X1 reached end of life caused some concerns about the longer term prospects for this enterprise storage system. Fortunately, as we announced last week, Park Place Technologies has added XtremIO to our long list of supported equipment from Dell EMC, IBM, HPE, Oracle, Cisco, Hitachi, and other OEMs. We are now quoting XtremIO contracts to start January 1, 2019.

We guarantee we can cut customers’ support costs in half. What’s more, our engineering team has done the hard work to understand the system’s architecture inside and out. That’s standard operating procedure for Park Place but even more important with this unique clustered array.

A Little Perspective on the XtremIO X1

The XtremIO X1 has had an interesting history. It went from early darling, rumored to be the VMAX replacement, to falling sales and unfounded speculation the product line might be discontinued.

To some extent, this is a tale of a start-up’s product finding a new home within the sprawling world of EMC, and XtremIO did face some challenges. After $25M investment by Israeli entrepreneurs returned a $430M payday upon acquisition, EMC needed to demonstrate that its expensive gambit would pay off with greater storage market dominance. The company clearly intended to meet the growing demand for all-flash arrays.

The problem was that XtremIO, Inc., had not brought out a single product, so EMC bought a development pipeline, not a proven winner. They needed to take it the rest of the way.

In doing so, EMC made some notable compromises.  Some reviewers complained about the lack of redundancy at the node level, an issue EMC addressed with dual controllers, multiple batteries, etc., to avoid having a single point of failure. Another issue has been the destructive nature of upgrades and system changes. For example, XtremIO was not be designed to scale-in. Removing a node from a cluster requires reloading all data from other media.

Great Performance that Stands Up Today

The XtremIO downsides do underscore the importance of excellent maintenance. Response to any first sign of failure must be addressed quickly to protect the node and its data. But there is good reason to stick with the system—namely performance.

The XtremIO architecture is based on coupled scale-out nodes called X-Bricks, and a proprietary, all-flash RAID architecture the company refers to as XDP. (And yes, if EMC was tied to the letter “V” with its VMAX and VNX, XtremIO laid claim to “X.”) The XDP RAID enabled a lowest-at-the-time, 8% overhead, which still looks pretty good today.

Although the capacity may be wanting—the XtremIO X1 scales up to 120 TB with the newer O/S—it can continue to do a good job for many database-intensive applications. Organization are also repurposing XtremIO X1s to the development farm, as they’re great for testing code ultimately aimed for mainframes.

The long and the short of it—a lot of IT organizations love the XtremIO X1 and they need to see the rather large investment in the system pay off. Affordable third party maintenance that includes engineering assistance, spare parts access, and remote monitoring with predictive AI built-in can ensure the XtremIO X1 remains a reliable contributor in any data center.

Need more information? Contact us.