To: Elmer Flugum who wrote (14641 ) 10/11/2002 12:41:40 AM From: Gus Read Replies (2) | Respond to of 17183 That article has a very sloppy premise, Len, because the writer doesn't even pay attention to the recent history of enterprise storage and its evolution in just the last 12 years. Once storage hardware started to look like a commodity, the game shifted to loading storage devices with software that was open enough to work with the other storage systems from competitors. And just exactly who started the trend of loading storage devices with software? EMC. EMC's first software product was SRDF, which came out in late 1994 after the 1993 WTC bombing. SRDF consisted of the functional layer (array-based microcodes or firmware) and the management layer (SRDF server). Before EMC disrupted IBM -- and its SLEDs (Single Line of Expensive Drives) -- with its RAID 1 boxes (Redundant Array of Inexpensive Drives) in 1990, practically every storage management task in the data center was performed by the mainframe and storage applications that run on the mainframe. Before EMC - in conjunction with the old HP and NCR -- disrupted the Unix and NT market with the Open System Symms, practically every storage management task was performed by the server and storage applications that run on the servers.Blinded by its success, however, EMC missed this turn in the road. Instead, it focused too long on hardware, forgoing a chance to take the lead in the new storage software game. Is this guy dumb or lazy? You be the judge. Here are the facts. In 1995, HOST-BASED backup and restore products accounted for more than 80% of the storage software market. In 1995 EMC had less than $25M in software revenues and less than 1% of that market. In 2001, HOST-BASED backup and restore products accounted for less than 40% of the storage software market. In 2001, EMC generated more than $1.4B in software revenues and captured more than 30% of that market, totally dominating the two fastest growing segments of the market: STORAGE INFRASTRUCTURE and ESRM. To reiterate, EMC went from less than $25M in 1995 to more than $1.4B in 2001 and yet somehow, EMC missed some imaginary turn that this writer doesn't even bother to explain? LOL. If we get lucky we may yet get to see this guy try to explain that the only storage software that really matters is the one that runs on the host because the abundance of processing power created by Moore's Law can not and should not be allowed to increase the intelligence of the storage array and the storage network.