To: cfimx who wrote (29179 ) 3/20/2000 12:41:00 AM From: rudedog Read Replies (1) | Respond to of 64865
twister - re: wildfire - let me first set some context here so that I don't get JC on my case. In 1995, when wildfire was first conceived, the world was a lot different than it is today. HP was the dominant player in the UNIX high end with the 9000 line, primarily the T500. DEC had a strong offering with the TurboLaser, an Alpha box which had about twice the T500's performance, but DEC had difficulty in selling outside of traditional DEC accounts. Both HP and DEC needed to refresh their high end UNIX products. SUNW had no high end servers but dominated the workstation space. DEC designed wildfire to be a replacement for the TurboLaser which would offer a more modular design, expandability to 64 processors with a roadmap to 256 processors, along with mainframe class I/O, all at a price performance point less than half of the current high end products. It featured a "switch fabric" architecture. DEC correctly determined that there would be an important market for high end UNIX machines a notch above where TurboLaser and T500 were playing, and that the first vendor to get there would get "architectural dominance" which would pull along huge sales of smaller systems. Typically for DEC they had great technical analysis and terrible execution. The product was initially hyped for a mid-1998 launch and word of the architecture and potential performance got to be fairly widespread by 1997. SUNW was suffering attrition in its workstation market and was hampered in its enterprise ambitions by the lack of a solid high end roadmap, and the flat stock price reflected the market notion that SUNW had capped out and would be pressured from above by HP, IBM and DEC and from below by Wintel. The market did not recon on two great moves by SUNW - Java and the acquisition of the technology which became the UE10000 from Cray which allowed SUNW to recreate themselves as an enterprise company. The competition also discounted SUNW, which gave them the opportunity they needed. HP did not produce a successor to the T500 which could capture the next notch up. IBM already owned that space with AS400 and S390, and had little incentive to replace that high margin business with lower margin UNIX boxes which would only help to breach the proprietary wall, so they also did nothing in that area. And DEC was so strapped by financial problems that they could not maintain momentum in any of their big R&D efforts. DEC sold their Alpha fab to Intel in late 1997 along with a lot of other assets including StrongARM. Internal cost cutting put wildfire in a virtual hold. Then CPQ bought DEC and everything DEC was doing got put on the back burner while CPQ reviewed what was going on. In the meantime, SUNW siezed the market opportunity, brought UE10000 to market, and created a whole line running from the UE3500 up, which although not seamless (the UE10000 had little in common with the rest of the line, including software) was still compelling enough to capture the UNIX high end. Wildfire, when it comes to market this summer, will be two years later than originally planned, and will find most of the chairs filled by UE10000 when the music stops. Recognizing the shift in the market, CPQ has repositioned DEC UNIX as a focused product line aimed at specific vertical markets. I am not completely clear on where they intend to go, but they pretty obviously don't intend to compete with SUNW as a general-purpose UNIX vendor. So although wildfire may have been a factor years ago in some SUNW product decisions, I doubt if it will be a competitive threat to UE10000 or its successors now.