Last year I made my first attempt at predicting what I thought would transpire with storage technologies and their use, integration, and trends related to server virtualization and cloud computing in 2011. I found sharing my views quiet enjoyable; however half the fun in making a prediction is following up to review them for accuracy.
Below are the seven predictions I made. My comments and a grade letter assigned to represent the quality of my prognostication. So without any further delay…
Cloud architectures will standardize on Network Attached Storage (NAS) as the de facto form of storage access. The deployment of NAS with VMware and other server virtualization technologies continued to grow in 2011, and while my use of the term de facto was a bit overstated consider some of the following area where we saw NAS growth.
a. VMware updated the vStorage APIs with vSphere to include NFS
b. Citrix stated NFS was the preferred storage protocol with XenDesktop deployments
c. Forrester Research published their second Storage Choices for Virtual Server Environments report in January of 2011. It reported that since 2009 NFS became the 2nd most deployed storage protocol. In 24 months NFS usage had grown 50% while the usage of block storage protocols was in decline.
d. VMware released the VMware Storage Array (VSA) that provides datastores only via NFS.
e. VMware Zimbra Collaboration Server added support for NFS
In retrospect, de facto might not have been the appropriate term to use in this prediction. However, the momentum around NAS continued to build in 2011.
Prediction Rating: B
Unified 10Gb Ethernet networking will increase in adoption, converging today’s disparate Ethernet and Fibre Channel networks.
This prediction has taught me to make a prediction quantifiable. I was hoping to have market share numbers by the time this posted, but alas this was not to be. With that said, I don’t expect many would challenge the growth of 10GbE in 2011. Partners like Emulex posted strong growth numbers with 10GbE and Cisco did exceptionally well with the Nexus platform. In addition FlexPod platform experienced massive success and each system is solely converged 10 Gb Ethernet.
Prediction Rating: B
The adoption of storage efficiency technologies like data deduplication with production data sets and applications will continue to increase.
It appears that I was a bit bullish on this prediction as well. While there was a significant increase in marketing collateral touting the use of storage efficiency technologies from the storage industry the adoption of with production workloads seems to be non-existent outside of NetApp’s customer base.
Maybe it’s the zero cost benefit that some storage vendors have introduced, where in customers need to purchase enhanced tiering/caching mechanisms comprised of SSD & net-new software in order to compensate for the negative performance attributes of their compression technologies.I believe it is fair to say that for this prediction I was naïve.
Expecting the storage industry to focus on enhancing their array platforms in order to sell customers less storage may have been too counterintuitive or too difficult for the industry to broadly adopt in 2011. Maybe 2012 we will see another storage vendor follow NetApp’s lead.
Prediction Rating: For my part I earned a D as I was off the mark. However I’d like to assign NetApp an A+ for continuing to drive these adoption of storage savings technologies out from the safety net of disk based backups and into production use cases!
The trend established in 2010 will continue with the announcement of additional strategic alliances and relationships dedicated to prevalidated cloud-based architectures like FlexPod.
Finally I nailed one, even if it was a no-brainer. While the stack market included Oracle and VCE 2011 saw it expand to include HP with their Converged Infrastructure and Dell who released vStart. I think I nailed this one, even if it was a lay-up.
Prediction Rating: A
Storage caching technologies, both hardware based modular expansion units and enhanced forms of caching software, will begin to gain acceptance as a primary form of storage access.
Storage tiering software, secondary array caches comprised of SSD, or PCI-E DRAM based caches installed in a host or an array (like FlashCache) were all the rage in 2011. For sake of this post I’ll skip the discussion as to why caching is becoming increasingly important; however, I do cover this need in depth in my forthcoming book, “Virtualization Changes Everything.” Storage caching dominated storage vendor presentations in 2011.
Prediction Rating: A
An increase in the number of applications that will integrate advanced storage array capabilities as a means of hardware offload, replacing constructs that historically were software based.
The market did experience an increase in the number of storage hardware integrations into software solutions; however, I don’t think the adoption occurred at the rate I had expected at the time I made the prediction.
Technology partners who led the way include VMware with their enhanced set of vStorage APIs and the backup solution providers CommVault and Syncsort. Both of these backup vendors integrated their backup, archival, and compliance tools with native capabilities provided by storage arrays to increase the scaling capabilities of their solutions. While there was some promising innovation in 2011, I feel that my prediction was too aggressive.
Prediction Rating: C
Software-based virtual storage arrays (VSA) will advance in their adoption, capabilities, and points of integration.
While VSA technology was introduced several yeas ago by LeftHand Networks (now HP) the architecture received mass visibility and market validation when VMware released their VSA in 2011. In addition NetApp released their ONTAP-v VSA thru an OEM agreement with Fujitsu and a number of storage startups promoting VSA and software-based arrays entered the market.
The largest surprise in this field was the absence of EMC who for years has distributed a virtualized version of their Celerra (now VNXe) array. I wouldn’t want to begin to speculate why EMC has choosen to remain out of the VSA market. VSAs received a solid endorsement as to their viability in 2011.
Prediction Rating: B
Well there it is, my first ever set of predictions and an accompanying review of each. I think its fair to say my prediction were fairly accurate even if sometimes the adoption rate was overstated. I hope you found the prediction concept enjoyable if not informative.
In retrospect I see where I could improve on my accuracy, so let’s see if I can’t apply the lessons learned to my future attempts at prognostication. I think it’s time to begin typing my thoughts for 2012 before January comes to a close. Cheers!