Monday, May 14, 2007

Cost vs. Benefit for Caching Appliances

Gary Orenstein at Gear6 sent me preview material with his press release for CACHEfx appliance launch.

I guess writing nice things about vendors once in a while has some benefits. May be, I will be crawling into doghouse, by the time Gary finishes reading this post. :-( But, somebody got to ask the hard questions.
Gear6 Unveils Industry’s First Terabyte-Scale Caching Appliances to Accelerate Data Intensive Applications

CACHEfx appliances support a baseline of 250,000 I/O Operations per Second (IOPS), 16 Gigabits per second of throughput, microsecond response time and scale linearly to handle millions of IOPS. The Reflex OS™ virtualizes appliance memory into a scalable coherent cache pool, optimizes data delivery through parallel I/O channels, and provides robust intelligent cache services.

CACHEfx centralized storage caching solutions are available now starting at $400,000.
My jaw dropped after looking at starting price tag of $400,000. My first reaction was "Damn, this thing is expensive!" The performance stats like 250K IOPS and 16Gbps throughput are impressive but let's be realistic how many customers can afford to pay $1.60 an IOPS to speed up applications, beyond hedge funds and stocks/options traders. I am looking forward to a ROI/TCO justification in near future.

How big is the market for caching appliances anyway? I hazard to guess that $1.6 an IOPS eliminates most web companies with intensive data access performance as well as a large portion of HPC market.

When and where can caching appliance threaten the parallel and clustered storage solutions?


  1. The trouble is that they are making something of a first. There's no well understood benchmark that shows the value add. They'll likely need to champion themselves with other nas companies to show their value. Like submit a SpecFS result with a FAS270, and show that their cache appliance helped it out perform a 6070. Which would be great for G6, but Netapp might have some issues. A TPC type benchmark with a $/iop would be killer, but most storage companies don't submit results there. Either route is a hard sell. Maybe on the higher end they can push a 6070 to even higher ranks of performance?

    The challenge is proving value. Most people don't shop for configs with lots of iops in mind, just $/gb. Once you start with a $/iop model, the value of g6 is much easier to find. And since it's not fixed memory attached to a storage platform, you get extra caching applied to any or all of your nas heads.

  2. I believe Gear6 differentiation lies in your statement Since it's not fixed memory attached to a storage platform, your get extra caching applied to any or all of your NAS heads.

    Intelligent adaptive load-based caching, if available, may be an attractive real value add for shops with dozens of NAS heads.

    I can see having extra 250K IOPS allocated as needed among an existing enterprise NAS network with large number of devices. Alternate choices come down to forklift replacement with faster NAS or parallel/clustered storage solutions.

    I don't think, Gear6 will find many takers for FAS270 delivering like BlueArc because of CACHEfx cost.

    But, Gear6 may be potential NetApp answer to losing deals to parallel/ clustered storage solutions and network of faster NAS solutions.

  3. Maybe I’m missing something, but it appears the Gear6 solution is designed to only support cache reads and not cache writes.

  4. Anonymous,

    My knowledge of Gear6 product is as good/bad as yours. You may want to pose your questions to Gary O at Gear 6.


  5. Check out what Gear6 has got going on now in terms of Cache Appliances -