Compellent CEO needs an Amazon Kindle

October 3rd, 2008 by Steven J. Schwartz

Image via Wikipedia

In a recent press release Phil Soran stated the following:

The LeftHand deal "…demonstrates legacy vendors’ approach to meeting the needs of the mid-sized enterprise market: buy someone else’s technology and try to fit it into an existing product portfolio, which leads customers to a rip-and-replace strategy as they grow," Soran said. "While this may be easier for the vendors than developing their own scalable architecture, it is certainly not better for customers."

     Seems like Mr. Soran either never read The Innovator’s Dilemma, or forgot what it was about.  The idea of “disruptive innovations” and large corporations purchasing new technologies rather then attempting to development internally, or the idea that a technology market that was previously not profitable or served will turn into a growth area for revenue and margin.

 

     The idea of HP buying into a storage software company like LeftHand Networks, and passing on the “Bridesmaid” Compellent, who is also firmly in the storage software market (Compellent’s product is a storage software cluster installed on a SuperMicro 3U server pair), is just poor luck, maybe Compellent should have used HP’s server platform rather then SuperMicro.  The idea of HP buying LeftHand Networks just shows the basic premise of Clayton M. Christensen.

 

     I think the scariest thing for the “Compellent”s of the world (small public storage up-starts) is the current state of the economy and trading markets.  Single product companies need to continue to expand product lines and execute during economic lows.  The best exit strategy for both LeftHand Networks and Equallogic, Inc. were to be purchased.  In recent financial news for Compellent, analysts are already lowering the expected target price for Compellent.

Related articles:

• Dennis Whittle: Innovators’ Dilemma in Philanthropy and Aid
• "Innovator’s Guide to Growth": Readable, Productive Prescriptions for Disruptive Innovation
• E-books: A Textbook Case
• Disrupting Class by Clayton M. Christensen
• HP to buy LeftHand Networks for storage virtualization, iSCSI

 

Technorati Tags: Compellent,SuperMicro,LeftHand Networks iSCSI,Compellent IPO,Equallogic Aquired,iSCSI Market

Posted in SAN and NAS, Start-up, virtualization | 3 Comments »

3 Responses

1. Chuck Hollis Says:
   October 3rd, 2008 at 12:13 pm

   Yep, on target again!!

2. John Says:
   October 3rd, 2008 at 5:20 pm

   Chuck left out a few points, like having 1 processor per 12 disks instead of one processor for 100s of disks. Clarion’s throughput bottlenecks are well published. And then there is the rip-and-replace growth strategy instead of just adding nodes. EMC’s usable capacity last I checked:

   Software Overhead, Cache Vault & 520 Byte Sectors 5%, RAID Parity (R5 4+1)20%, Spare drives(EMC best practices 1 out of 30)3.3%, Recommended Snap Space Reserve 20%, add remote data Replication (redundancy across sites) with MirrorView/S 50% and usable capacity is 37%.

   This is worse than LeftHand configured with multi-site redundancy. The bottom line is that the EMC guys think they know it all, when in fact they are tied to a Data General FLARE “ball-and-chain”, which is good, because we will just continue to eat their lunch, and in a big way going forward.

   John

3. John Says:
   October 3rd, 2008 at 5:34 pm

   With regards to Compellent, sooner or later customers will get past the illusion of Data Progression and realize that this is the same old legacy controller head and disk shelf architecture that everyone and their mother sells. And what’s worse is that they don’t have a product family like EMC and NetApp do. I don’t know about you, but I wouldn’t want an indeterminate amount of my critical database bits sitting on unreliable SATA disk with no RAID 6 (this is what happens when Data Progression does its thing.) They will say, “just use RAID 10.”; there goes your savings.

   Compellent also has one of the slowest storage systems in the industry because of all the metadata management, software RAID overhead etc, etc, and I challenge them to prove otherwise. They are the kings of marketing, but reality will hit them sooner or later.

Leave a Comment