Arun Taneja, Founder
On March 8, HPE announced it will be acquiring Nimble for approximately $1B. This acquisition comes on the heels of the acquisition of SimpliVity for $650M that closed on February 17, 2017. So what is HPE up to and what does it all mean for the industry? We can write a book on this topic but we will restrain ourselves and briefly focus on the major ramifications for now.
Let’s start with state of HPE and the industry, as it stood before these acquisitions. In terms of storage, HPE has enjoyed a solid run with 3PAR. One could even argue HPE’s storage portfolio has only consisted of the 3PAR family of products for several years. Of course, there are other products in the family but the mainstay has been 3PAR. Even when its competitors had to buy or build an all-flash array from scratch, 3PAR products upgraded well to an all-flash format (primarily due to its ASIC-based architecture). When all hell was breaking loose in the traditional array revenue, HPE was holding its own, even growing slowly, on the back of 3PAR. In fact, it gained market share in 9 out of 11 quarters and maintained its #2 position in total storage (internal and external), as of Q4, 2016. But the low end was suffering and MSA was only selling into existing accounts, and at a declining rate at that. StoreVirtual wasn’t carrying its weight. StoreOnce, while growing, was not doing so at a rate the product merits. As a full-line supplier, HPE needed to strengthen its offerings, especially at the low end.
On a different front, HyperConvergence was catching on and customers were beginning to buy Nutanix and SimpliVity (and later, VxRail) and a plethora of other products, to simply their infrastructure deployments/management, while reducing costs. And while HPE had a #1 presence in servers, its customers were beginning to go to HyperConvergence players whose products were based on non-HPE servers. Given that the growth rates in the HC space was (is) heady and servers are an important portion of the solution, HPE was bound to lose its server dominance.
Buying SimpliVity solved several major issues for HPE. It gave them a presence in the fast growing HC space and it allowed them to hold on to the server revenue. Even beyond that, it allows HPE to keep the revenue associated with storage internal to the HC solution. Buying SimpliVity only for $650M made the transaction that much sweeter.
Buying Nimble solves a different set of problems for HPE. While its implications are a little more complex, the simplest way to view the Nimble acquisition is to view it as filling the low-to-mid range of traditional storage arrays space. It should quickly replace MSA across the board and keep those customers within the HP fold (many were looking for alternative solutions and 3PAR was an overreach). We even see Nimble replacing most, if not all, StoreVirtual product line, over time. This would be especially true if Nimble could be presented as a software-defined storage product that can run as a VM, either on-prem or in the cloud (he recently announced, but not yet shipping Nimble Cloud Volumes seems to suggest exactly that). Yes, StoreVirtual does have a scale-out architecture but in the low to mid-range, Nimble provides enough performance and storage capacity to not run into scalability issues. In a nutshell, Nimble stops the bleeding on the low end and gets the growth engines moving again.
The main issue that HPE has to deal with is how to manage the mid-range, where both Nimble and 3PAR can perform well. We suspect Nimble may be more cost effective and could put pressure on 3PAR from below. But 3PAR can scale to higher levels. Many questions remain unanswered at the moment. How does each perform in an all-flash array version? We don’t know yet. This will be the most difficult challenge, we believe, for HPE to manage. Having said that, we do believe that having an overlap in functionality is better than having a gap in functionality. As HP tests out the Nimble product it will become clear where it excels and where it doesn’t. That will become fodder for positioning the two products appropriately.
Along with Nimble comes InfoSight, the gem that uses a mound of near-real time metadata from thousands of its customers to assist in diagnostics, analytics and prediction. InfoSight today only works with Nimble storage but Nimble was already working to extend its functionality to non-storage products (servers, networking, maybe even applications). That means it should be able to apply the generic code to other storage products as well. HPE can now combine its own StoreFront Remote product functionality with InfoSight to gain a serious advantage in the market over Dell EMC, IBM, NetApp and others.
To be sure, there is a lot of work to be done to make HPE a powerhouse in storage again. Both 3PAR and Nimble are traditional storage arrays. We think HPE needs storage offerings for the new world of cloud and software-defined. Its presence in the object-storage space is weak. Its NAS offerings are still practically non-existent. It needs to have a greater presence in the space occupied by StoreOnce.
But we believe the two acquisitions made in the last two months show that HP is moving aggressively to backfill functionality it was badly missing. Much more needs to be done. We were concerned that the re-organization had all but paralyzed HP. These acquisitions show that not to be true. The industry needs HPE to be a strong presence in the infrastructure market. It’s good to see HPE get its mojo back!