News
 

Bookmark and Share

(4) 

Intel Corp. this week disclosed certain details regarding the company’s octa-core Intel Xeon processors code-named Beckton that are projected to be released either in very late 2009 or sometime in 2010. Apparently, modular architecture of the chip as well as so-called “cache and core recovery” technology will allow Intel to disable defective cores and caches within the new chip and still sell the product.

It is well known that the bigger the chip is, the lower is its yield, hence, potential yield of very large Intel Xeon processors with eight processing engines may be rather low. IBM once said that the initial yield of Cell processor was from 10% to 20% unless there is redundant logic put inside, which automatically boosts manufacturing costs. In order to tackle the issue, Intel has developed “cache and core recovery” scheme for its chips and will use it for the first time in Intel Xeon processors with eight cores aimed at multi-processor servers, reports Ars Technica web-site.

According to Intel’s presentation at IEEE International Solid-State Circuits Conference (ISSCC), eight-core Xeon processors will be able to operate its processing engines at different clock-speeds and voltages (from 0.85V to 1.1V) or even disable them when they are not needed, whereas the other logic’s clock-speeds and voltages will be fixed.

Moreover, the “cache and core recovery” technology will allow Intel to disable caches or cores independently in case they are defective. This will allow Intel to create processors not only with eight cores out of the Beckton dies, but also make chips with sever-core, six-core, five-core or any other amount of cores that makes sense. This means that if a Beckton silicon has one core and one cache domain malfunction, it will not be scrapped, but sold as a micro processing unit with lower amount of cores.

In fact, presently available Intel Xeon 7400-series processors have from four to six cores, which means that even now Intel has some control over the number of functioning engines within its monolithic processors.

The new enterprise Intel Xeon processor based on code-named Nehalem micro-architecture featuring 8-cores (16-threads due to HyperThreading support) has 2.3 billion transistors and is made using 45nm process technology. The chip that is known under Beckton code-name will have four point-to-point quick path interconnect links to connect to other processors as well as system I/O operating at up to 6.4GT/s. Like all Nehalem-class chips, this processor for expandable systems will feature built-in memory controller, which will have four channels. The processor will require a new platform with LGA-1567 sockets.

Tags: Intel, 45nm, Beckton, Nehalem

Discussion

Comments currently: 4
Discussion started: 02/16/09 06:06:36 AM
Latest comment: 02/19/09 08:30:39 PM
Expand all threads | Collapse all threads

[1-3]

1. 
Wow, Intel following AMD's lead yet again. AMD has been doing this ever since Phenom X3 was released.
0 0 [Posted by: mamisano  | Date: 02/16/09 06:06:36 AM]
Reply
- collapse thread

 
Not really surprising that intel just now starts this, since theyve only been using the monolithic quad cores and larger processors for a few months now.
0 0 [Posted by: Eighty-Sixed  | Date: 02/16/09 07:02:27 PM]
Reply

2. 
Considering how much ass kicking Intel has been doing with their processors, I think they can be forgiven for taking a bit out of AMD's playbook
0 0 [Posted by: digitalrurouni  | Date: 02/16/09 07:50:20 PM]
Reply

3. 
^ If it was not for AMD we would be still using Intel single cores and paying a premium on top too
0 0 [Posted by: alpha0ne  | Date: 02/19/09 08:30:39 PM]
Reply

[1-3]

Add your Comment




Related news

Latest News

Monday, July 28, 2014

6:02 pm | Microsoft’s Mobile Strategy Seem to Fail: Sales of Lumia and Surface Remain Low. Microsoft Still Cannot Make Windows a Popular Mobile Platform

12:11 pm | Intel Core i7-5960X “Haswell-E” De-Lidded: Twelve Cores and Alloy-Based Thermal Interface. Intel Core i7-5960X Uses “Haswell-EP” Die, Promises Good Overclocking Potential

Tuesday, July 22, 2014

10:40 pm | ARM Preps Second-Generation “Artemis” and “Maya” 64-Bit ARMv8-A Offerings. ARM Readies 64-Bit Cores for Non-Traditional Applications

7:38 pm | AMD Vows to Introduce 20nm Products Next Year. AMD’s 20nm APUs, GPUs and Embedded Chips to Arrive in 2015

4:08 am | Microsoft to Unify All Windows Operating Systems for Client PCs. One Windows OS will Power PCs, Tablets and Smartphones