In one of my latest reports, “Key Criteria for Evaluating Enterprise Block Storage,” I wrote that analytics is one of those features that make a difference and has an impact on overall system usability as well CTO.
System-level monitoring and analytics
The number of vendors providing this kind of functionality is now pretty large but the level of functionalities they offer can be totally different and, sometimes, different products from the same vendors have different tools!
Some of the tools available in the market are very mature. They provide insight on the state of the infrastructure, help to understand trends, and planning for future infrastructure changes. It is also very important to note that these analytics systems become even more relevant for system management when integrated with call-home support functionalities, helping to predict potential issues and remediate them before they become critical. In this category, you can find tools like HPE InfoSight, NetApp ActiveIQ, or Pure Storage Pure1 Meta.
Storage analytics at the system level makes life easier but what happens when things get really complicated?
Infrastructure-level monitoring and analytics
Larger organizations have complex infrastructures, this is a fact. Usually, they do not have a single storage provider. A deeper and broader vision of what is actually happening at the infrastructure level is needed. Additionally, in some cases, they need application visibility as well.
In this case, the system-level approach is no longer enough. The risks are to get an incomplete picture or underestimate some of the issues present in the stack. But how can you get complete visibility on complex network and storage systems?
Fortunately, there are tools that are more focused on providing a view of the entire infrastructure than the single storage system and this gives a better view of what is happening in every layer of the stack starting from the application down to the storage array. The flip side here is that these kind of products can’t support every system in the market and, most of the time, can’t give you information on the array internals. This means that you can’t avoid having both system- and infrastructure-level monitoring and analytics for your infrastructure as both are critical.
One of the most important characteristics of infrastructure-level monitoring and analytics is ease of use and visualization. In fact, if you know there is a problem but are unable to visualize and address it quickly it becomes very hard to justify the presence of these products in your infrastructure.
A good example of what I’m talking about can be found in Virtual Instruments, the demo they presented at the last VMworld was quite impressive and this is only a part of it. In fact, I suggest you check their integrations with App Dynamics to have a complete picture of what their solution can do.
Small organizations can tremendously benefit from data storage analytics. Understanding how the storage system works can save a lot of time and money.
As soon as the infrastructure grows and becomes multivendor and complex, we need an approach that is more holistic and this is where infrastructure-level tools come into play.
Unfortunately, these two approaches are not alternatives to each other, they are both needed to understand what is happening in your data center at any time and avoid any potential service disruption.
If you want to better understand the key criteria to evaluate enterprise block storage I suggest you check my report on the topic here on GigaOm.