Bill Fischofer said:
EMC realized many years ago that the real challenge and opportunity is in storage management. In the enterprise space storage demand is growing at a mind-boggling pace. The net result is that either storage management productivity grows apace or else you're faced with doubling and redoubling the size of your IT staff every year. Even if you could find the people the staffing expense would be prohibitive. That's why EMC has invested and continues to invest a huge portion of their R&D budget in storage management software. I've preached this message for many years on the EMC thread but in the storage business there is only one metric which matters: How many Gb can be managed per IT staff person. By this metric EMC remains streets ahead of anyone else, which is why those who only focus on the boxes and their pricetags miss the bigger picture.
To which you replied (in part):
If computer device industry at large can combine important pieces of the puzzle in commodity form (iRAID, iNIC, iNP, and BSD Unix all within the framework of I2O), Dell-like OEMs will grab the low end of server appliance devices, including storage. The management software won’t be nearly as capable as EMC’s, but adequate given huge reductions in product costs. In time, management software available throughout the Linux/BSD Unix free source community, combined with proprietary developments, will be adequate for a growing number of companies.
I agree with you. It appears that EMC may be facing the classic "disruptive innovation" as described in Clayton Christianson's The Innovator's Dilemma. In fact, in this case I think that competitive products in the near future might be not only cheaper than EMC's, but might also very quickly become as capable or more capable. The way I see this happening is via software from Veritas--the hands-down leader in storage management software. Over the last few years, Veritas has been growing steadily both in terms product functionality and financial strength. They have clearly beaten their competition in this space [which is mainly Legato, CA/Arcserve, and IBM's Tivoli Storage Manager (formerly known as ADSM)], and they are a pure software play. In a nutshell, I can see where:
Veritas Software + Storage Hardware Built from Commodity Parts might very quickly equal or exceed the functionality of EMC at a lower cost--particularly in non-mainframe environments. If WIND can position themselves to get a piece of this, I'll be delighted. :-) |
