back to article Symmetrix and the death of monolithic arrays

EMC's Symmetrix V-Max, announced today, represents the death of monolithic storage arrays and the rise of combined scale-up/scale-out modular arrays built from commodity components tied together with ASICs, very clever software and geographical distance cluster interconnects. It's also a tribute to 3PAR's T-Class InServ design …

COMMENTS

This topic is closed for new posts.
  1. Tom Cox

    RapidIO supports very large systems

    RapidIO 1.3 supports large complex systems and RapidIO 2.0 is approved and now in systems designs.

    RapidIO is today capable of handling much large systems then 256, and has a new generation of switches which are at 2X the speed, Adds more virtual channels and has data management capabilities. Wireless and Wired Communications systems use RapidIO and in some cases have done so for 10 years. The military has strongly enbraced RapidIO for the same reasons as EMC, as Burke states.

  2. Eddie Johnson
    Coat

    If you try really hard...

    ...you could probably jam a few more buzzwords in that first paragraph and make it totally incomprehensible. Oh, wait, nevermind.

  3. Jan

    looks like a DS8X00 on steroids...

    ´bout time someone copied IBM.

    I mean, it´s not like the DS8k is exeptionally fast anyway [1]... But this whole Symmetrix thingie looks like a combination of a DS8k and a SVC.

    [1] otoh, its also not exceptionally expensive, and I´d choose a good midrange array in my datacenter anytime above an enterprise arry in catalogue.

  4. Tim Parker
    Stop

    Rumours of my death...

    "represents the death of monolithic storage arrays"

    No it doesn't. Please stop being a mouth-piece for storage vendors announcements Chris - it's getting pretty dull.

  5. Dwayne

    IBM SVC Tax

    What's the deal about SVC?

    V-Max is intelligent storage engines that communicate over a redundant fabric - Rapid IO. No intelligence in the fabric - Just really fast switches.

    SVC is the hardware tax you pay to IBM to allow rebranded netapp, ibm, and whoever elses array IBM wants to sell this week to be managed as a single array.

    Th idea of dropping in a new v-max array onto the floor and it federates itself into the virtual storage environment connected by a very high speed fabric is a very compelling vision of the future for data migration, business continuity, and disaster recovery.

  6. Anonymous Coward
    Anonymous Coward

    The deal about SVC

    > What's the deal about SVC?

    Well, the deal is you can take a truckload of 'el cheapo' arrays, plug them almost braindead into a SAN and still get away with it since the SVC will provide all the management and redunancy to place enterprise workloads onto them.

    Persistent rumor claims there are shops out there which have a pair of DS6k high performance/pricetag arrays and around two dozen of HP MSA2212/2312 (and a bunch of other stuff) arrays spread evenly across two datacenters quite some miles apart.

    If you´d plug all those into an SVC cluster you´d be able to mirror drives without OS interaction, could evacuate arrays going down for maintenance and move demanding workloads temporarily (or permanent) onto the expensive storage without the clients ever noticing.

    Well, the users call when their jobs suddenly run in 1/4th of their previous runtime. And then they call again when I move^H^H^H^Hwould move them back onto the MSA... ;)

    Really unscrupolus Sysadmins woud even run Raid accross several cheap boxes and get amazing performance out of dirt-cheap arrays.

    Even with the SVC licenses for far less than proper arrays would cost.

This topic is closed for new posts.