Disclaimer: I recently attended Storage Field Day 8. My flights, accommodation and other expenses were paid for by Tech Field Day. There is no requirement for me to blog about any of the content presented and I am not compensated in any way for my time at the event. Some materials presented were discussed under NDA and don’t form part of my blog posts, but could influence future discussions.
For each of the presentations I attended at SFD8, there are a few things I want to include in the post. Firstly, you can see video footage of the Nimble Storage presentation here. You can also download my raw notes from the presentation here. Finally, here’s a link to the Nimble Storage website that covers some of what they presented.
Nimble Storage have been around for a while now, and I put up an enthusiastic post about their InfoSight product after hearing about it at Storage Field Day 6. The cool thing is they’ve been working hard at improving what was already an impressive offering.
InfoSight is still awesome
Rod Bagg provided an overview of InfoSight. Nimble have spent a lot of time working on what they call “Operational Intelligence”. They asked a few pointed questions of their products:
- “In a connected world why can’t vendors proactively monitor customer deployed systems?”
- “With modern data analytic tools can vendors predict and prevent problems before they occur?”
Nimble’s design philosophy for InfoSight is as follows:
- Be intuitive – present use cases and not just data
- Be prescriptive – provide specific recommendations for immediate action
- Be predictive – estimate future needs based on current and past learning
Rod then shared some “fun facts” about the statistics they’ve been collecting:
- 1000s of unique sensors recording operational data each second,
- 30-70M data points collected from every array every day,
- >20000000 heartbeats every week
- 200B log events
- by-day view of every config element of every array
- lifetime data from day 1
- rich install base
- data from 1000s of arrays for 5+ years
- dedicated team of data scientists on support staff
- advanced analytics techniques,
My favourite thing about all of this is the idea that you can use InfoSight as a key part of protecting your investment. If your vendor has access to statistics about 1000s of deployed systems, why wouldn’t you use them to help you with the following exercises?
Nimble has access to “1000s of system-years” of real-world data. They also have an understanding of workloads correlated to resource consumption. In this fashion you can aim to understand the exact configuration before you make your purchase.
Predictive Capacity Recommendations
InfoSight also provides you with the ability to perform continuous storage capacity prediction, and plan your storage purchase in advance. This is invaluable when working in environments where budgets are allocated at fixed points in time.
As Nimble has a whole bunch of data on workload and behaviour across a number of systems, they can help you with the working-set analysis for cache and CPU optimisation. With this information you’ll also be able to understand the optimal cache, scale-up and scale-out requirements before you deploy the system.
Closing Thoughts and Further Reading
I was excited about InfoSight when I first saw it in action, and I remain an enthusiastic advocate for this approach to understanding your storage environment. I love the idea of taking a lot of the guesswork out of platform sizing, in addition to making good use of the available data. While the Nimble Storage hardware isn’t for everyone, I encourage you to have a look at them if you’re in the market for a hybrid array, simply by virtue of the fact that the InfoSight product has the potential to provide a valuable insight into what your storage is doing on a daily basis.