Bookmark and Share


Intel Corp. does not have plans to introduce brand-new Xeon Phi coprocessor cards every year, like it does with its central processing units, but the company does intend to release new accelerators from time to time. In addition to already announced Xeon Phi 3100-series products due in the first half of the year, Intel will also unveil new 5100- and 7100-series coprocessors starting this May.

According to a leaked Intel roadmap, the world’s largest chipmaker intends to release Xeon Phi 5120D, 3120A, 3120P, 7120P and 7120X coprocessor products sometimes in May – July, 2013, which is in line with the plan to bring Xeon Phi 3100-series to the market in the first half of the year. Exact specifications of the co-processors are unknown, but all of them are based on Knights Corner chip. The Xeon Phi 3100-series was supposed to deliver over 1TFLOPS peak double precision performance, whereas Xeon Phi 5120D, 7120P and 7120X coprocessors are supposed to be faster. At present, Intel does not have exact shipment date for the new products, but expects them to become generally available from May 1, 2013 to July 31, 2013.

The Intel Xeon Phi coprocessor 3100 family will provide great value for those seeking to run compute-bound workloads such as life science applications and financial simulations. The Intel Xeon Phi 3100 family will offer more than 1TFLOPS double-precision performance, support for up to 6GB memory at 240GB/s bandwidth, and a series of reliability features including memory error correction codes (ECC). The family will operate within a 300W thermal design point (TDP) envelope. The Intel Xeon Phi coprocessor 3100 will have recommended customer price below $2000.

The Intel Xeon Phi coprocessor 5110P, the only Xeon Phi available today commercially, features 60 cores with 4-way simultaneous multi-threading technology and 512KB L2 cache per core, provides additional performance at a lower power envelope. It reaches 1.01TFLOPS double-precision performance, and supports 8GB of GDDR5 memory at a higher 320 GB/sec memory bandwidth. With 225 watts TDP, the passively cooled Intel Xeon Phi coprocessor 5110P delivers power efficiency that is ideal for dense computing environments, and is aimed at capacity-bound workloads such as digital content creation and energy research. This processor has been delivered to early customers and featured in the 40th edition of the top500 list. The Intel Xeon Phi coprocessor 5110P has recommended customer price of $2649.

Based on the Intel Many Integrated Core (Intel MIC) architecture, Intel Xeon Phi coprocessors delivers high performance for highly parallel applications.

More than 50 manufacturers are designing solutions based on the Intel Xeon Phi coprocessors, including Acer, Appro, Asus, Bull, Colfax, Cray, Dell, Eurotech, Fujitsu, Hitachi, HP, IBM, Inspur, NEC, Quanta, SGI, Supermicro and Tyan.

Tags: Intel, Xeon, Xeon Phi, Knights Corner, 22nm, MIC


Comments currently: 9
Discussion started: 03/26/13 08:52:15 PM
Latest comment: 07/13/16 11:11:29 AM
Expand all threads | Collapse all threads


It's a bit pointless to list FLOPS rate when it doesn't even support modern instruction sets, thus it needs to waste more cycles to do the same level of calculations of a normal processor. Considering each 'core' is a Pentium I core with a few changes, but lacking critical components that a modern processor uses, there isn't many programs that these can effectively run. Considering the price tag, I fail to see why anyone would want to buy these. At least today's graphics cards have much better support and processing speed, along with cheaper prices.
1 2 [Posted by: mmstick  | Date: 03/26/13 08:52:15 PM]
- collapse thread

Ooh... looks like the troll crawled out of the woodwork!

Go look at some independent benchmarks before you insult the Phi: http://goparallel.sourcef...eon-phi-shocks-tesla-gpu/

Oh, your rant about Pentiums is completely wrong too. The heavy-lifting in the Phi is done by 512-bit vector units that are included in each of the on-board cores. Despite what you might think, the Pentium didn't have those. Saying that the Phi == Pentium is like saying that AMD's or Nvidia's latest GPUs aren't any better than my TI-82 because they all rely on transistors to do anything....

0 1 [Posted by: chuckula  | Date: 03/27/13 05:06:47 AM]
Pointless comparison since you can't run any real scientific application on this. It has just tiny 6 Gb of RAM on board, while simple FEM/FDTD projects for RF start from 16 Gb of RAM. Nuclear simulations need much more. So, this is a useless product just to point out Inte's presence on scientific market. I think they are stuck with Xeon E5 performance top floor and can't move further where real market demands lie.

Back in times of real coprocessors on 286/386/486 platforms they had full direct access to system memory through CPU. These toys can only pick up memory windows selectively, but the direct RAM access is slow as hell. Quad XEON platforms are much more productive and single i7 platform is more productive for single threaded calculation (many integral equation approaches can't be paralleled)
0 0 [Posted by: Gena Gennadievich  | Date: 04/27/13 07:07:23 AM]

1 TFLOP may be fast, but if don't want to cut on the costs you should buy Radeon HD 7870 which has 1.7 TFLOP and costs less than 200$. It will save you something around 1800$ on initial purchase and then some more bucks when your electricity bill will come.

Don't misunderstand me - I run all my computers on Intel CPUs because right now they are faster, more energy efficient and cooler than any other CPUs, but if you need to compute a lot of stuff and needs to do it really fast and cheap, there is nothing worse than buying Intel or Nvidia GPUs to do the job.
0 0 [Posted by: knedle  | Date: 03/31/13 05:39:04 AM]
- collapse thread

Yes, but you are comparing single precision with double.

0 0 [Posted by: Kimberly Peacock  | Date: 05/08/13 06:20:15 AM]

for god,s sake... produce more... lower the price....
0 0 [Posted by: Franc Lee  | Date: 06/04/13 08:00:36 AM]

deadly needed...for dft_2d large size x87 long mode... where gpus are useless...
0 0 [Posted by: Franc Lee  | Date: 06/04/13 08:04:31 AM]

great design x86+x87 (float64, float80)... just a little step will reach it,s highest bless: forget soft double-double, just hardware enable it,s 512 vec. float128.... will shy all gpu cards away....
0 0 [Posted by: Franc Lee  | Date: 06/04/13 08:14:57 AM]


Add your Comment

Related news

Latest News

Wednesday, November 5, 2014

10:48 pm | LG’s Unique Ultra-Wide Curved 34” Display Finally Hits the Market. LG 34UC97 Available in the U.S. and the U.K.

Wednesday, October 8, 2014

12:52 pm | Lisa Su Appointed as New CEO of Advanced Micro Devices. Rory Read Steps Down, Lisa Su Becomes New CEO of AMD

Thursday, August 28, 2014

4:22 am | AMD Has No Plans to Reconsider Recommended Prices of Radeon R9 Graphics Cards. AMD Will Not Lower Recommended Prices of Radeon R9 Graphics Solutions

Wednesday, August 27, 2014

1:09 pm | Samsung Begins to Produce 2.13GHz 64GB DDR4 Memory Modules. Samsung Uses TSV DRAMs for 64GB DDR4 RDIMMs

Tuesday, August 26, 2014

10:41 am | AMD Quietly Reveals Third Iteration of GCN Architecture with Tonga GPU. AMD Unleashes Radeon R9 285 Graphics Cards, Tonga GPU, GCN 1.2 Architecture