Home | Opinion | Storage Old, New and Past Due? (Updated)
Storage Old, New and Past Due? (Updated)

Storage Old, New and Past Due? (Updated)

4 Flares Twitter 0 Facebook 0 Google+ 4 StumbleUpon 0 Buffer 0 LinkedIn 0 4 Flares ×

It’s been an interesting day of contrasts in the world of storage, one that shows storage is a diverse and wide ranging segment of IT.

The Old

Tape has been part of the discussion on the twitterverse and despite everyone’s best attempts, is not dead yet.  Tape and backup may not be seen as cool  - but data protection is an essential requirement of sustainable data management and tape still provides the one of the most cost effective methods of data protection and of course archive.  This has happened because tape continues to innovate.  Tape drive speeds and media capacities continue to push upwards to meet demand, driving the effective cost per GB down and so keeping tape a player in the long term data retention market.  Tape will be around for a long time to come.

The New

Flash storage is all the rage and today Pure Storage announced they have finally gone GA with their all-flash storage arrays.  They have also produced pretty funny video, taking a side swipe at traditional storage arrays using spinning disk.

YouTube Preview Image

Whilst this was a clever piece of marketing, it’s more useful to understand how this technology is implemented and why flash in a traditional array was only a stopgap.  At the recent Storage Field Day, Pure Storage presented a technical deep dive of their architecture, explaining some of the thoughts that led to their second generation array – available from today.  There was some pretty amazing detail presented, including a discussion on maintaining I/O latency when an SSD decides to falter.  The Pure Storage array can choose to recreate the data from parity rather than wait for the I/O to complete and so maintain low latency.  This is how solid state should be designed into storage arrays.  Here are the videos from the Pure Storage presentation.

YouTube Preview Image YouTube Preview Image YouTube Preview Image

 

Past Due

The Register carried two articles today discussing EMC VMAX, which is due for a refresh and expected to be announced at EMC World next week.  The first talks about EMC scaling VMAX to 4PB of storage and/or 3,200 drives.  This is a huge capacity to store in a single platform and represents a massive amount of information to keep in a single chassis.  Symmetrix will probably go down as one of the most successful and pivotal storage arrays in history, however I think it is coming close to the end of it’s useful life because:

  • flash will be the dominant technology for high performance applications
  • bulk capacity can be done cheaper and easier
  • vendors are building technology towers, not centralising storage in the way they did 10 years ago
  • intelligence is being pushed up to the hypervisor
One of the issues with placing such a large quantity of data into a single chassis is the ability to migrate to and from the array, especially when the device is due to be decommissioned.  Perhaps this is one of the reasons why EMC has also chosen to implement storage virtualisation, which was the subject of The Register’s other article today.  Yes, it’s true, EMC are finally admitting Storage Virtualisation is cool and HDS and IBM were right all along.  One of the easiest ways to migrate data in and out of large arrays is to virtualise.  What’s ironic is the way EMC (and their Symmetrix strategist Barry Burke) have parodied the idea of storage virtualisation in so many blog posts.  Here’s just a few to savour:
The fact is, EMC had to have some technology built into VMAX to enable migration.  Otherwise, building 4PB arrays creates the customer a world of pain.
I have no details, but if the technology EMC is using here is RecoverPoint, then it’s hardly a native solution, but will be a sticking plaster before the arrival of XtremIO finally puts Symmetrix out to pasture.

 

 

About Chris M Evans

  • psl

    “…before the arrival of XtremIO finally puts Symmetrix out to pasture.”

    Which will be an interesting dynamic at EMC, since Symmetrix is under the Enterprise Storage Division (ESD), and XtremIO was acquired by the Unified Storage Division (USD). USD is home to VNX and Isilon. There is limited cooperation between the divisions. They simply aren’t incentivized to cooperate.

    • http://thestorageanarchist.com the storage anarchist

       Actually, we work across divisions more than you apparently know.

      And as expensive as a 4PB VMAX 40K may be, I’d be interested to see how it would compare to a 4PB All-flash array. My calculator still has the $/GB of SLC or eMLC NAND at several multiples more than both 15Krpm “performance” and 7200rpm nearline “capacity” drives. I thus believe that most all-flash arrays will soon add support for HDDs – the economics arent’ there: not all data has the need for speed.

      Then, to compete for the enterprise storage requirements, this new batch of all-flash contenders will need to implement HA scale-out, snaps, clones, remote replication, FIPS-140-2 encryption, multi-tenancy, access security, etc. – a rather long list of features that few (if any) currently deliver.

      I’m not saying that they can’t do all that, just that low-latency performance alone doesn’t make a product the Symmetrix Killer.

      Now, admittedly it is my job (I am the Chief Strategy Officer for EMC’s Enterprise Storage Division for those that don’t know), but I’m of the opinion that EMC can match pure flash arrays on performance faster than they can match Symmetrix on features and proven track record.

      • chrismevans

        Barry, some good points.  I’m interested to know why EMC therefore have decided to purchase an all-flash vendor if you’re implying Symmetrix is still the array of choice.  Are you saying the XtremIO purchase was for niche solutions, and if so doesn’t it make the purchase effectively more expensive?

        Chris

        • http://thestorageanarchist.com the storage anarchist

           Chris –

          I did not mean to imply that Symmetrix was the array of choice for all market segments, only that it unlikely that any all-flash array can pose a serious challenge vs. the high-end arrays simply because they are fast – “enterprise storage” is not defined by performance alone.

          Many make the mistake of over-simplifying the storage market into a single set of requirements serving the entire bread of storage solutions, when in fact there are several (often overlapping) segments to the market with different (often overlapping) requirements. EMC long ago abandoned any pretense that the storage market was a singularity, and instead began investing organically and inorganically to deliver best-of-breed products optimized around different segments. Many of EMC’s larger customers use products from multiple EMC product lines, while the vast majority of customers select (and/or start with) a single product.

          My comments about all-flash arrays are based upon my experiences with several of EMC’s larger customers. These often realize that if the data is so critical that it needs the lowest-possible latency, it almost always also requires the same data integrity, security, availability, and reliability characteristics that they demand from their enterprise arrays. For these, a faster Symmetrix will probably be a better fit than would a hot-rod all-flash arrays with unproven enterprise characteristics.

          I suspect there are other segments of the market for whom storage performance is paramount, and they will be willing to pay a premium for the best performance even if it means that it will take a while to recover from a data center outage or disaster.

  • Pingback: Storage Field Day 1: The Links

  • http://thestorageanarchist.com/ the storage anarchist

    Chris –

    VMAX has been delivering Symmetrix seamless migration solutions for more than 5 years: Open Replicator/Live Migration and the more recent Federate Live Migration serve to provide tech refresh migrations with only a brief out (using OR/LM(, or withe ZERO downtime (when moving from any DMX sold since 2003).

    FTS is not RecoverPoint – it is a native implementation. RecoverPoint splitter functionality has also been integrated, and in fact the two work well together.

    And there is much, much more to FTS than sweating assets…which I can’t talk about just yet.

4 Flares Twitter 0 Facebook 0 Google+ 4 StumbleUpon 0 Buffer 0 LinkedIn 0 4 Flares ×