As you may have heard, in early March, Hewlett Packard Enterprise announced its intention to acquire independent hybrid and all-flash array maker Nimble. On April 17, this purchase was completed and the company is now 100% owned by HPE. In countries where Nimble has previously been introduced, Nimble products are now available through the Hewlett Packard Enterprise channel. In our country, this process will take longer, but we can expect that by November, Nimble arrays will occupy their niche between the older MSA and 3PAR 8200 configurations.
Along with the integration of manufacturing and sales channels, another challenge for HPE is to leverage Nimble InfoSight software that goes beyond storage alone. By , InfoSight is the industry's leading predictive IT health analytics platform that other vendors are trying to replicate. HPE currently has an equivalent - , however both IDC and Gartner rated Nimble significantly higher in their 2016 All-Flash Array Magic Quadrant. What are the differences?
InfoSight is changing the way storage infrastructure is managed. It can be quite difficult to determine the source of problems that may arise in the “virtual machine - server - storage system” bundle. Especially if all these products are supported by different vendors (I remind you that in the case of HPE, ). It would be much easier for the user if a comprehensive analysis of the state of the infrastructure was carried out automatically at all levels of IT through which business application transactions pass, and the results were already issued in the form of a ready-made solution. And preferably before the problem arises. Nimble InfoSight software does just that, delivering unique results: data availability in fact, on entry-level systems, and automatic prediction of potential problems (including those outside the storage system) with the implementation of preventive measures in 86% of cases. Without the participation of the system administrator and calls to the support service! In general, if you want to spend less time maintaining your information system, I recommend taking a closer look at InfoSight.
How is this done?
One of the key differences of the NimbleOS operating system is the greater amount of diagnostic data available for analysis. Thus, instead of standard logs and system state metrics, a huge amount of additional information is collected. The developers call the diagnostic code "sensors", and these sensors are built into each operating system module. Nimble's installed base has more than 10000 customers, and tens of thousands of systems are connected to the cloud, which currently collects 300 trillion array metrics over the years of operation, and analyzes millions of events every second.
When you have so many statistics, all that's left to do is analyze them.
It turns out that more than half of the problems that cause business application I/O slowdowns are , and other storage-only vendors simply cannot adequately understand the service case in most cases. By correlating array data with other diagnostic information, you can find the real source of problems all the way from virtual machines to array disks. Here are some examples:
1. Performance diagnostics is quite a challenge for a complex IT infrastructure. Analyzing log files and metrics at each level of the system can be time consuming. Based on the correlation of many indicators, InfoSight is able to determine where the slowdown occurs - on the server, in the data network or in the storage system. Perhaps the problem is in the neighboring virtual machine, perhaps the configuration of the network equipment is incorrect, perhaps the server settings should be optimized.
2. Invisible problems. A certain sequence of indicators forms a signature that allows you to predict how the system will behave in the future. Over 800 signatures are monitored by the InfoSight software in real time, again allowing problems to be detected outside of the array. For example, after upgrading the storage operating system, one of the customers experienced a tenfold drop in performance due to the peculiarities of the hypervisor. Based on this case, not only was a patch released, but another 600 storage systems were automatically prevented from having a similar situation, since the signature was immediately added to the InfoSight cloud.
Artificial Intelligence
Perhaps this is too strong a phrase to describe the work of InfoSight, but nevertheless, advanced statistical algorithms and predictions based on them are a key advantage of the platform. The algorithms used by the platform include autoregressive forecasting models and Monte Carlo simulations, which allow predicting "random", as it may seem at first glance, events.
Data on the current state of the infrastructure allows you to make absolutely accurate sizing for the modernization of the information system. Since the deployment of new components, InfoSight receives data for further analysis, and the mathematical model becomes even more accurate.
The platform is constantly trained on the installed base that has been built with customers over the years of Nimble's existence, and it is trained to make maintaining systems - now Hewlett Packard Enterprise - a simpler and more understandable task. The number of 3PAR arrays alone currently in service at customers far exceeds that of Nimble. Accordingly, InfoSight's support for 3PAR will create an even more complete picture for the statistical analysis of IT infrastructure indicators. Of course, modifications to 3PAR OS will be required, but, on the other hand, not everything that is built into InfoSight is unique to this platform. Therefore, we are waiting for news from the joint development team of Hewlett Packard Enterprise and Nimble!
Materials:
Source: habr.com