To: stockman_scott who wrote (4689 ) 12/23/2002 2:15:51 AM From: David A. Lethe Read Replies (3) | Respond to of 4808 I'm not familiar with MonoSphere, but my 10,000 foot view is to stay away from these guys. Block-level virtualization is certainly something that the marketplace is looking for, especially with managed SRM. Hardly "unique" in the marketplace as their press release below states. One big problem is their desire to blend virtualization with their own SRM package. They are way under capitalized to do this. Once they actually have a product to ship, and start running into interoperability problems with 3rd party storage, HBAs, operating system issues, HBA driver software and patches, then they'll find themselves in a situation where they will only be able to qualify a small number of vendors. If they go after the short list of EMC/HDS, then they'll find that those vendors already have qualified SRM and virtualization vendors, so they won't get the time of day from them to get API info and/or do quals. They have 50 employees, and the got 9M of financing. Their test lab will cost millions to buy hardware and get the right skills to just do testing and support. They need much more money before they can properly support customers. The release also mentioned they would consider OEMing to BMC, Tivoli, or CA. It is extremely unlikely these vendors would want to OEM their software. (Maybe if they had some interesting patents, then that would be reason to buy the company, but certainly not OEM their software. The problem with block-level virtualization is backup and recovery. Basically you are limited to snapshots. That is because a file/filesystem agnostic mechanism has no concept of file names and where a particular chunk of data is ... so you gotta still buy expensive backup software to facilitate selective data recovery. The pricing model gets less attractive. Now .. where I see this going is distributed block-level virtualization companies will continue to bleed, and then when the switch vendors start shipping virtualization blades that fit inside the switches, these guys and the others in this space will have serious financial problems. Also nothing here about specifics on patents. Tamir Ram, their architect that came out of Auspex learned the hard way about not owning their own IP. For years, they (Auspex) licensed their engine from another company (CrossRoads or CrossStor, don't remember). When that company got bought by Auspex's competetor, they told Auspex they wouldn't renew the contract so Auspex was forced to write their own core engine. They almost didn't recover from that. David