New characteristics of co-processors Xeon Phi

For several years, developed Intel Larrabee graphics cards have changed its destiny, and now proudly referred to coprocessors Xeon Phi. The June announcement gave a very rough idea of ​​their characteristics and applications (highly specialized servers and workstations). Now some details will be revealed.

Bboosters Xeon Phi will be available in versions with 57, 60 and 61 core. There are in general 64 computational units, but some of them will be disabled. Depending on the number of cores the cache size varies from 1.8-1.9 MB first level and 28-30,5 MB second. The volume of onboard GDDR5 memory is 3, 6, and 8 GB.

Intel Xeon Phi device will operate on a low clock rate from 600-1100 MHz for the lower 57-core models with 3 and 6 GB of GDDR5, to the 630-1050 MHz or more for the older 60 — and 61-core board with 8 GB of RAM. Effective memory frequency will be in the range of 5-5.5 GHz. Support for dynamic overclocking Turbo Boost will be provided by reasonable balance of …

Palit Intros GeForce GTX 750 KalmX Graphics Card

Palit, one of the world largest manufacturers of graphics cards, came up to the market with one more edition to the GeForce GTX 750 series – the GeForce GTX 750 KalmX. The model employs etalon PCB with PCIe 3.0 x16 interface and a two-slot passive cooling system. The latter accommodates a massive nickel-plated heatsink, consisting of 43 thin fins, and two heatpipes that connect the heatsink with a copper base. The unique model is already available in German online-stores for 114.9 EUR.

Plait GeForce GTX 750 KalmX is geared up by GM107-300 GPU (28 nm, Maxwell architecture) with 512 CUDA cores. Its base/efficient clock frequency makes up 1020 MHz/1185 MHz. The chip is accompanied by 2 GB of GDDR5 memory with 128-bit interface; its operational clock frequency approximates 5010 MHz.

The list of external interfaces, which are located on the rear panel, includes a couple of DVI ports and a single mini-HDMI port. One should take into consideration that the card is void of an …

Processors Intel Haswell to Rival Accelerator NVIDIA Tesla

Intel Corporation has the intentions to introduce DDR4 memory into their server processors of Haswell series. The launch of such technological approach is planned for the 2014, however the developments in this sphere have already started. Intel’s server solutions will become a part of Grantley platform and the samples of their new series will allegedly be announced next autumn. The preview of new processors makes it clear that some of the chips will be distinguished by the increased heat generation performance which will reach 160 W.

Hence, 16-core processors Haswell-EP will rival parallel computing accelerator NVIDIA Tesla but not AMD server solutions. Intel is going to provide various working frequency options and options with different number of active cores. At the same time, TDP will fluctuate within ranges of 160, 145, 135, and 120 W. In its turn, accelerator NVIDIA Tesla, which is based on GK110 processor, consumes approximately 300 W.

In accordance with the …
Get computer-assisted driver search with our brand utility *
* The Utility requires installation. Upon installation and running DevID Agent will detect
which devices require drivers and which drivers require updates.
The Utility doesn't contain viruses or malicious code
Registered at: 07 September 2012, 15:41
Last visit: 17 November 2012, 20:12