Connect with us
Xilinx calls Alveo U55C its highest performing accelerator card Xilinx calls Alveo U55C its highest performing accelerator card

Components

Xilinx calls Alveo U55C its highest performing accelerator card

Published

on

Calling itself a leader in adaptive computing, Xilinx unveiled the Alveo U55C Data Center Accelerator Card and a new standardized API clustering solution for mass FPGA deployment at the SC21 supercomputer conference. The Alveo U55C accelerator is said to “deliver superior performance per unit of power in supercomputing and database workloads.” It scales easily with the Xilinx HPC clustering solution.

The novelty, specially designed for the tasks indicated above, is the most productive accelerator card in the entire history of Alveo, surpassing its predecessors both in computational density and in HBM2 memory, equal to 16 GB.

The advantages of the accelerator include high parallelism of data pipelines and optimized movement of data along pipelines. The card comes in a full height half length (FHHL) form factor. It occupies one slot, and its power consumption does not exceed 150 watts.

The above API-based clustering solution leverages RoCE v2 and bridging combined with 200 Gbps of bandwidth to form the backbone of Alveo’s network, which competes with InfiniBand in performance and latency. MPI integration allows supercomputer developers to scale the pipeline processing of data coming from the Xilinx Vitis unified software platform using Alveo. Leveraging existing open standards and frameworks, hundreds of Alveo cards can be bundled together, working with shared workloads and memory, regardless of server platforms and network infrastructure.

Software developers and data scientists can take advantage of Alveo and adaptive computing by leveraging the high-level programming tools in which Xilinx has invested heavily. Major AI frameworks such as Pytorch and Tensorflow are supported, as well as high-level programming languages ​​such as C, C ++ and Python.

Alveo U55C card is already available. To familiarize yourself with its capabilities, you can use the FPGA-as-a-Service cloud services.

Click to comment

Leave a Reply

Your email address will not be published.

Components

Rumor: Nvidia Stops Production of GeForce RTX 3080 GPUs After GeForce RTX 3080 Ti Drops in Price

Published

on

Rumor Nvidia Stops Production of GeForce RTX 3080 GPUs After

A steep rise in the value of major cryptocurrencies triggered a cascade of events that resulted in the collapse of prices for new video cards. Many models have fallen in price, including top ones like the GeForce RTX 3080 Ti: with a recommended price of $1,200 (in the US), it is offered by the largest American retailers, Newegg and Amazon, for less than $1,000.

Rumor: Nvidia Stops Production of GeForce RTX 3080 GPUs After GeForce RTX 3080 Ti Drops in Price

A small price gap between the GeForce RTX 3080 Ti and the GeForce RTX 3080 with 12GB of memory has led Nvidia to suspend shipments of GPUs to partners for the latter. So wrote one of the Twitter users familiar with the situation. It is impossible to confirm or refute these words, but in general such a development of events can be considered quite probable.

Continue Reading

Components

Up to 8 XE cores and 6 GB of memory: the first information about the Intel Arc Pro A50/A40 graphics cards appeared

Published

on

Up to 8 XE cores and 6 GB of memory

The database of the SiSoftware Sandra test package has fresh data on the upcoming Arc A40 Pro and Arc A50 Pro graphics cards. These are graphics accelerators for workstations that are designed for engineering tasks.

Up to 8 XE cores and 6 GB of memory

The A50 Pro has a full ACM-G11 GPU with 8 Xe cores (128 execution units/1024 FP32 cores), while the A40 Pro is likely to get a stripped-down version of the chip. The clock frequency is 2.45 GHz.

Both cards have 6 GB GDDR6 memory, indicating a 96-bit memory bus. In this regard, both models are similar to the Arc A380 mobile card, which has almost the same configuration.

1656342768 632 Up to 8 XE cores and 6 GB of memory

Thus, the company continues to prepare its entry-level products. When high-performance gaming and professional solutions based on the ACM-G10 GPU will appear, it is not reported. It is also not yet clear whether the Arc A40 Pro and Arc A50 Pro cards will be available in retail or whether they will appear only as part of workstations.

Earlier, Intel introduced its first gaming graphics card for $150.

Continue Reading

Components

Named the likely date of the announcement of the mobile processor Qualcomm Snapdragon 8 Gen 2

Published

on

Named the likely date of the announcement of the mobile

Qualcomm has updated its 2022 product launch roadmap. And there, among other things, they indicated the likely date for the announcement of the flagship Snapdragon 8 Gen 2 processor. It is expected to be shown on the first day of the Snapdragon Summit event, which will be held from November 14 to 17.

Named the likely date of the announcement of the mobile

Qualcomm also announced the Snapdragon X70 5G modem that will be included with the Snapdragon 8 Gen 2. It is credited with download speeds of up to 10 Gbps, as well as new advanced features for working in fifth-generation networks.

1656325903 885 Named the likely date of the announcement of the mobile

The processor (SM8550) will be manufactured using TSMC’s 4nm process. It will use a three-cluster layout. The composition is expected to include one super-powerful Cortex X3 core, two productive Cortex-A720 and A710, as well as three energy-efficient Cortex-A510. An Adreno740 graphics accelerator is also expected.

It is assumed that the first smartphone based on the Snapdragon 8 Gen 2 processor will be Xiaomi 13, its presentation is expected before the end of 2022.

Continue Reading

Most Popular