Join Newsletter
Forgot
password?
Register
Trusted Business Advisors, Expert Technology Analysts

Taneja Blog

Taneja Blog / Software Defined/Virtualized Infrastructure

The Convergence-ware Battle

It is interesting to see this space emerge as we continue to see substantial solutions take shape over the course of a couple of years now.  We have considerable functionality in many of these products now, and it has been an exciting exercise to put several of them through the paces in the lab, as well as do this higher level assessment of key primary storage capabilities.  It has also been quite enjoyable to get some more personal value from some of these products by injecting serious capability into our lab storage.  This in particular has been a hoot.  While TGLabs does an excellent job of simulating specific exercises that take place in day to day businesses and enterprises, I am the first to acknowledge that it is far from a real operational environment. No matter what's in the lab, I don't wear a pager, and I don't get calls at night.  Even so, many of these technologies are certainly ready for prime time in many businesses, and they've so significantly eased our storage tasks that I frankly can't imagine living without them.

One of the things that I've concluded as we approach a hiatus in some of our VSA work, is that there are bound to be some wrinkles for bigger customers as more practitioners turn to these software solutions (for reasons of agility and efficiency typically).  You see, the software does a lot, but at the end of the day, the practitioner is taking on a lot too.  

We're now taking on the responsibility of determining how many spindles are good for these solutions, how much SSD capacity we need, how the spindles should be plumbed through our systems (especially when they may be used in a very different way than the server or RAID card manufacturer may have first expected), and how multiple systems should be connected together for fault tolerance or scale out. Only 7 to 10 years ago this used to be the core focus of our attention when choosing a storage array and what we paid big bucks for - remember those marketing wars focused on plumbing, ports, and spindle counts behind enterprise arrays?  Now we're saying we're going to take this on ourselves.  Hmmm.

Thinking about this, it seems on the surface that this is proof of value in hyperconverged solutions that can deliver storage and compute scale-out.  But the reality is many remain dubious about the seeming lockin with these fairly proprietary solution stacks - if your vendor goes boom in the night, how do you maintain your growth strategy?  I think there's a middle ground too, and HP is doing a pretty good job of making this middle ground highly attractive with their StoreVirtual VSA technology and pairing it with HP hardware.  If you can simplify the software architecture, provide good visibility, and then make some good pairings with the right hardware, the benefits of simplified and field-ready design might be harnessed alongside a bit more openness and freedom from lock-in.  

So what I’m saying is that we’re in the midst of a convergence-ware battle, and it isn’t just old-way vs new-way.  The new way is being actively defined, and some things are going to work and not.  A multi-faceted battle is going on among the new stuff - between software storage, hyperconvergence, and this middle ground of traditional hardware with enhanced software-driven capabilities. Clearly we're getting considerable value as the battle drives more innovation and an on-going paradigm shift around how we build our data centers, but the battle is pretty nuanced, and if you’re looking for long term data center investments, you should be paying attention to the really, really big battlefield, and thinking about what is really going to leave a lasting influence.  

The thing is that this really big battlefield is bigger than the historic markets where small innovative disruptive upstarts come in and really change things.  This new battlefield is tied to multiple data center domains and this makes it a huge market with more players than ever before (think about the last time you saw just the server industry get disrupted, and what the disruptor had to look like to make it happen).  This market historically isn’t about tremendous tumultuous and sudden change, but rather about small inflection points and gradual evolution.  So in this convergence market, it is my bet that we’re more likely to see gradual change take place in the intersection of symbiotic technologies that advance the integration and capabilities of the whole data center rather than witness upheaval and breakthrough leadership by small innovators.  

With that in mind, as the industry raves about disruptive technologies, what gets missed is the middle ground, where the lasting change is most likely to get built.  In my opinion, it is likely to be innovation from the big vendors that drives change and really advances capabilities in this market.  Folks like HP toting around solutions that have been field proven for 5+ years are likely to be the ones who drive that lasting change.  There's fully baked stuff out there that is changing the way we do storage.  For me, it might just be our labs.  For you, it is bigger.  Drop me a line, and tell me your story:  jeff.boles@tanejagroup.com">jeff@tanejagroup.com.

  • Premiered: 12/23/13
  • Author: Taneja Group
Topic(s): VSA Virtualization Storage TGLabs SSD Flash HP StoreVirtual StoreVirtual VSA

Comments

There are no comments to display. Scroll down to leave your own!

 

Leave a Comment

You must be logged in to comment. Click here to log in or register if you don't have an account.