I recently was invited to Cloud Field Day in the Bay area. While in attendance the group heard from many different companies. One of them was HPE Nimble Storage. I really didn’t know what to expect for this presentation, but as the discussion took form the strategy being used was quite interesting.
In your on-premises infrastructure design there was a time when you couldn’t imagine using anything other than the local storage that came with the servers you were purchasing. While this was a long time ago, we are not embarking on a new time where the infrastructure our servers run on is in the cloud within a provider offering where they once again call the shots on the storage we use. Shouldn’t we have a choice? HPE Nimble Storage has a cloud-based SAN-like option for organizations to use with their cloud-based servers. Interesting, right?
Challenges with block storage
When HPE shared their challenges on why the storage that a cloud provider may offer cause performance challenges I initially thought hmm. I can’t say that I ever experienced a storage issue in the cloud, but then I paused again deciding that my use cases likely didn’t fit the bill. So what are the challenges? HPE stated the following: “Durability and features (1 in 500 failure rate), and lack of data services”, cloud lock-in (data mobility is hard, data egress costs), and Black Box Penalty – Limits visibility.”
One huge value add to their platform is their InfoSight analytics option. The value with this comes from the fact that it will allow you to become more proactive to hardware issues instead of reactive.
From a security perspective, they are not analyzing your corporate data, but what they are doing is analyzing millions of sensors per second on their offering. Their goal is to protect their customers both latencies and/or any other potential problems that may arise.
Here are the statistics they shared with us on the success they are seeing with InfoSight:
- Predicts and prevents problems 86% of the time
- Email notification to the customer with a case
- 99.9999% Measured availability
- 54% of cases solved outside of storage
- Only level 3 support engineers are hired, level 1 and 2 issues are automated/prevented
- 91% of all Nimble arrays in the world send their data for analytics
So what does the customer gain with this level of analytics? Issues that become identified are remediated and rolled out across the platform to ensure that other customers do not have the same issue.
I think that this offering by HPE Nimble Storage is very interesting, but I do not see companies running out and wanting to implement this at the beginning of their cloud journey. Customers are going to first see if they are getting what they need from their cloud offering for their servers before spending even more money on top-tier storage to add to their cloud offering.
While this type of offering is great, it may be better suited for direct use with one of the major cloud players. For example, when I decide to go with cloud company A for my servers, this option should be right there within the deployment platform with a partner.