AMD Launches FirePro S9170 Accelerator - 32 GB GDDR5 Memory and Grenada's 5.24 TFlop Compute Power

After launching several cards for the consumer marketplace, AMD is dorsum in the professional person marketplace with an update in the course of the FirePro S9170 Accelerator. Regarded as the successor to the FirePro S9150, the S9170 comes with the Grenada cadre architecture that features an insane 32 GB of GDDR5 memory and massive amounts of compute aimed at DGEMM (Dense General Matrix) and server minded workloads along with full OpenCL 2.0 support.

AMD FirePro S9170 Packs 32 GB GDDR5 Retention and Grenada XT GPU - Aimed at Double Precision Workloads

The AMD FirePro S9170 is an update to the FirePro S9150, aiming at server and HPC markets. Both cards are based on the Hawaii XT GPU cadre with 44 Compute Engines that pack 2816 stream processors, 176 TMUs and 64 ROPs. The departure lies within the power optimizations and efficiency improvements which were recently added to the Radeon R9 390X and Radeon R9 390, in specific the Grenada GPU core which is a revised version of Hawaii.

Compared to the 900 MHz clock speed on the S9150, the S9170 comes with a clock speed of 930 MHz that pushes the FP64 (Double Precision) performance to 2.62 TFlops compared to 2.53 TFlops on the S9150 while the FP32 (Single Precision) performance is leveraged to 5.24 TFlops compared five.07 TFlops on the S9150. The major difference betwixt the S9170 is that it has a max board power of 275W compared to 235W of the S9150. In that location is an optional 235W version available for other suitable workloads. The card is fed power through a 8+half-dozen Pin power configuration that is constitute on the back of the PCB.

Moving onwards, the S9170 gets a big update on the retentiveness forepart. AMD has configured this top tier HPC bill of fare with 32 GB of GDDR5 retentiveness which is the near institute on any graphics card. The 2nd card to feature a high memory is the Tesla K80 with 24 GB GDDR5 VRAM but even that doesn't match what AMD has on board their HPC accelerator. HPC cards practice actually require large pools of VRAM for professional usage to render intense tasks, simulations and models.

"Nosotros accept been developing a fully-parallel computational tool based on the AMD GPU heterogeneous calculating platform and OpenCL," said Omid Mahahadi, co-founder and director, Geomechanica Inc. "This tool accurately captures the circuitous physics of massive mines plus oil and gas fields quickly and reliably. Thanks to the impressive 32GB of retentiveness of the new cards, we expect to run computations on massive information structures containing tens of millions of information elements. The combination of rapid double-precision operations with the large retentivity capacity enables authentic, detailed, and reliable computations. A like operation using CPUs would probable crave much higher capital letter and maintenance costs. Moving forward, we plan to take advantage of the recent features of the OpenCL 2.0 open API to farther enhance the performance of our software."

"What makes AMD's GPUs peculiarly well suited for HPC boils down to iv things: very high compute functioning; leadership in operation-per-watt; the largest available working memory size; and our embracing of open standards, including the most recent revision of OpenCL," said Niles Burbank, senior product manager for professional graphics at AMD.

The AMD FirePro S9170 offers unparalleled GPU compute performance and power efficiency with the post-obit benefits:

• With up to ii.62 TFLOPS of meridian double precision performance, the AMD FirePro S9170 is the fastest unmarried-GPU server carte du jour available for DGEMM heavy workloads, delivering up to 40% more performance than the competitive solution;
• Support for 40% better double precision performance, while using 10% less ability than the competition;
• The AMD FirePro S9170 is the industry's first server GPU with 32GB ultra-fast GDDR5 on-board retention and features a 512-bit memory interface for 320 GB/s of retention bandwidth;
• Equipped with 32GB of GDDR5 memory, the AMD FirePro S9170 GPU can accelerate retention-intensive applications and process larger and more computationally complex workloads with ease; and
• The AMD FirePro S9170 GPU features 33% more retention than the competitive GPU, helping to improve overall workload speed and system responsiveness, particularly when working with big amounts of data. via AMD

While AMD and NVIDIA accept fast HPC options, information technology should be noted that neither visitor has withal launched a HPC focused accelerator based on their latest, top-tier cores. NVIDIA made clear that due to lack of FP64 blocks on the Maxwell cards, they won't offering products in the Tesla market place and Kepler volition be offered in 2022 as the merely Tesla options until 2022 when NVIDIA introduces their Pascal GPU architecture with second generation HBM retentivity. AMD is also taking a similar approach as their Fiji GPU may not be making way to the FirePro S-Series market since its a gaming focused chip and a refocus on the server marketplace is expected in 2022 with FinFET GPU offerings.

AMD FirePro S9170 Graphics Card Gallery:

AMD FirePro S9170 Specifications:

	AMD FirePro S9170	AMD FirePro S9150	NVIDIA Tesla K80	NVIDIA Tesla K40
GPU Cadre	Grenada XT	Hawaii XT	2 x GK210	GK110B
GPU Engines	44 Compute Engines	44 Compute Engines	xiii SMX	15 SMX
GPU Stream Processors	2816 SPs	2816 SPs	4992 CUDA	2880 CUDA
GPU TMUs/ROPs	176/64	176/64	208 / 48 x2	240 / 48
GPU Clock	930 MHz	900 MHz	Upwards to 875 MHz	Up to 875 MHz
VRAM	32 GB GDDR5	16 GB GDDR5	12 GB GDDR5	24 GB GDDR5
Memory Clock	5.0 GHz	v.0 GHz	5.0 GHz	6.0 GHz
Memory Passenger vehicle / Bandwidth	512-scrap / 320 GB/southward	512-bit / 320 GB/south	384-chip / 480 GB/s	384-bit / 288 GB/s
FP64 / FP32 Compute	ii.62 / 5.24 TFlops	2.53 / 5.07 TFlops	2.91 / 8.74 TFlops	ane.43 / four.29 TFlops
TDP	275W (235 Optional)	235W	300W	235W

Source: https://wccftech.com/amd-launches-firepro-s9170-accelerator-32-gb-gddr5-memory-grenadas-524-tflop-compute-power/

Posted by: crozieracked1968.blogspot.com

Share this post