If you’ve followed my blog for some time now you might have read an article or two about a product Nimble Storage created called InfoSight. Hey, if not, we’re all busy and I’ll forgive you this time…
Nimble Storage InfoSight - Empowering The Storage Administrator!
Replication Made Easy with Nimble Storage - Part II - Planning with InfoSight
If you’d like to learn more about InfoSight visit the Nimble Storage website, but here’s a small bit about the coolness of InfoSight, taken from the Nimble Storage website.
At the heart of InfoSight is a powerful engine comprised of deep data analytics applied to telemetry data gathered from Nimble arrays deployed across the globe. More than 30 million sensor values are collected per day per Nimble Storage array.The InfoSight Engine transforms the millions of gathered data points into actionable information that allows customers to realize significant operational efficiency through:
- Maintaining optimal storage performance
- Projecting storage capacity needs
- Proactively monitoring storage health and getting granular alerts
I was in a meeting today with one of the Nimble Storage Sales Engineers and he told me an InfoSight story that I thought was so cool that I asked him to be a guest blogger. With that, I’d like to introduce Mark Antonaccio. Take it away Mark!
Thanks Neil! This story starts back in May of this year. I had just started as an SE at Nimble, and right away I experienced the power of predictive analytics and the value it could bring to my customer. I’ll leave this customer unnamed, but I can say that they are a global leader in the payments industry. They have multiple Nimble storage arrays, and they use InfoSight to proactively manage the performance and capacity of each. Here’s the story:
Our customer received an email from Nimble support that notified them of a new software release. I met with them to discuss upgrading their arrays to the latest version, informing them that it would provide new functionality, including per-volume encryption and cache pinning. I also opened a ticket with Nimble support to determine if any additional steps would need to be taken as part of the upgrade. Here’s what we found out - the array was actually blacklisted and was currently not eligible to be upgraded. InfoSight showed that we needed to correct a latency issue due to cache underprovisioning and possible sequential I/O contention. So right away, InfoSight saved us from moving forward with the upgrade and possibly inducing performance problems in a production environment. Now, we needed to dig deeper and find the root cause. Our wonderful Nimble support folks were on the case, and they quickly found the cause of the cache issue. Log volumes were being placed into cache with no benefit, reducing the available cache space for DB volumes that could utilize it. These log volumes were misconfigured within the Nimble array. There was a straightforward, non-disruptive, and simple fix that could be performed by one of our capable Storage Architects. We wrote a script that automated the creation of a new performance policy that did not cache data, and applied it to all of the log volumes. We then saw the cache utilization drop, clearing the way for the upgrade to take place.