Intel: x86 won't encroach on Itanium

TalkBack 1 of 36:

Next »

• Thread View
• Flat View

Intel's uncertain processor road-map (or the lack of a road...)

``x86 in the driver's seat
Gartner's Reynolds believes that x86 servers have captured the attention and momentum of the server market. "The small servers have destroyed the growth opportunity for big iron," he said.''

It is clear that the dynosauric x86 'architecture' has already extended its useful life to a point that a further middle or long term extension is untenable. x86 and its 64-bit extensions may survive at the desktop for 3-5 years MAX but this is not true at ALL in the server space. The server platforms require a solid, extensible and scalable architecture which NO x86 is capable of providing. Intel knows this very well and it knew it back in the middle 90's when it designed the IA64 for its more serious systems.

One place that x86 may survive longer is in the cluster platforms where massive numbers of lower capacity processors can be made to cooperate at a considerable communication costs. The downside to this is that there are very few s/w environments and applications that can leverage the larger numbers of independent nodes. Only scientific computing apps can leverage this environment. There is some progress in DB and distributed file systems but clusters suffer from communications bottlenecks, unless fitted with the expensive high-speed interconnects.



``At the same time, Itanium has the potential to grab a bigger share of the market that remains, at the expense of Sun Microsystems' UltraSparc and IBM's Power, he said.''

The big iron server space requires billions of investments in processor and chip technologies and a long-term planning. Itanium just barely started catching up with the latest multi-core technologies and it is still a hog of memory bandwidth. It appears that Intel didn't know how to design a processor for a large SMP. IBM, HP, SUN and others do know the ramifications. It seems to me that Itanium was designed for the high-end workstation market and not the server one.



``So far, Itanium machines have made a modest dent in the server market. Server customers purchased $1.6 billion worth of Itanium machines in 2004, up from $448 million in 2003, Reynolds said.''

Because it is lacking a roadmap for its evolution into an SMP processor. Intel has been very lucky that SGI and HP made this processor something of value for the presetnt SMP technology. How about the future though?

Intel is stuck extending the popular x86 but it knows that the only long-term solution is the IA64 but it has to decide to develop the missing technologies.

Posted by: michael-t   Posted on: 03/08/05 You are currently: a Guest | Members login | Terms of Use

Alert moderator to an offensive message

Subscribe to this discussion via Email or RSS

SponsoredWhite Papers, Webcasts, and Downloads

• Using NIC Teaming to Achieve High Availability on Linux Platforms Dell Network interface card (NIC) teaming is one method for providing high ... Download Now
• Storage Capacity Management for Oracle Databases Quest Software Do you have sufficient space capacity to support your growing Oracle ... Download Now
• Building the Virtualized Enterprise with VMware Infrastructure VMware VMware virtualization software has been adopted by over 120,000 enterprise ... Download Now