Half-precision floating-point format

AMD Showcases Leadership Cloud Performance with New Amazon EC2 Instances Powered by 4th Gen AMD EPYC Processors

Retrieved on: 
Friday, August 18, 2023
Cloud, Central Information Technology Office, Amazon Web Services, Amazon Elastic Compute Cloud, AMD, Female gendering of AI technologies, Intelligence, M7, Computational chemistry, DDR5, Forecasting, DDR4, Amazon, AMD EPYC, Half-precision floating-point format, VNNI, EC2, HPC, DTN, DNT, Weather, AWS, Memory, TrueCar, Video game

SANTA CLARA, Calif., Aug. 18, 2023 (GLOBE NEWSWIRE) -- Today, AMD (NASDAQ: AMD) announced Amazon Web Services (AWS) has expanded its 4th Gen AMD EPYC™ processor-based offerings with the general availability of Amazon Elastic Compute Cloud (EC2) M7a and Amazon EC2 Hpc7a instances, which offer next-generation performance and efficiency for applications that benefit from high performance, high throughput and tightly coupled HPC workloads, respectively.

Key Points: 
  • —Amazon EC2 Hpc7a instances, powered by 4th Gen AMD EPYC processors, deliver up to 2.5x better performance compared to Amazon EC2 Hpc6a instances—
    SANTA CLARA, Calif., Aug. 18, 2023 (GLOBE NEWSWIRE) -- Today, AMD (NASDAQ: AMD) announced Amazon Web Services (AWS) has expanded its 4th Gen AMD EPYC™ processor-based offerings with the general availability of Amazon Elastic Compute Cloud (EC2) M7a and Amazon EC2 Hpc7a instances , which offer next-generation performance and efficiency for applications that benefit from high performance, high throughput and tightly coupled HPC workloads, respectively.
  • “For customers with increasingly complex and compute-intensive workloads, 4th Gen EPYC processor-powered Amazon EC2 instances deliver a differentiated offering for customers,” said David Brown, vice president of Amazon EC2 at AWS.
  • Amazon EC2 Hpc7a instances are designed for tightly coupled high performance computing workloads and deliver 2.5x better performance compared to Amazon EC2 Hpc6a instances.
  • With these new instances, AWS customers can continue to take advantage of the excellent performance, scalability, and efficiency offered by AMD EPYC processors.

FriendliAI Launches Public Beta of PeriFlow Cloud

Retrieved on: 
Thursday, July 20, 2023
Internet, Data Management, Technology, Artificial Intelligence, Software, Cloud, MPT, LLM, OPT, GPT, Organization, Traction, Llama, AI, GPU, BLOOM, Video game, T5, Half-precision floating-point format

FriendliAI , a leading generative AI engine company, is proud to announce the public beta release of PeriFlow Cloud .

Key Points: 
  • FriendliAI , a leading generative AI engine company, is proud to announce the public beta release of PeriFlow Cloud .
  • This powerful platform empowers users to run PeriFlow , an engine for generative AI serving, within a managed cloud environment.
  • FriendliAI also offers PeriFlow as a container solution, named PeriFlow Container, which has gained considerable traction among companies for LLM serving.
  • We are incredibly excited to see the innovative services users will develop with their generative AI models, powered by PeriFlow Cloud.”
    The public beta version of PeriFlow Cloud is now available.

AMD Reimagines Cloud Performance with 4th Gen AMD EPYC Processors with AWS

Retrieved on: 
Tuesday, June 13, 2023
Cloud, Amazon Web Services, Amazon Elastic Compute Cloud, AMD, Female gendering of AI technologies, Fourth, McLaren M6A, M7, Amazon, AMD EPYC, Half-precision floating-point format, VNNI, EC2, CPU, EPYC, DNT, AWS, TrueCar, Video game

SANTA CLARA, Calif., June 13, 2023 (GLOBE NEWSWIRE) -- Today, during the “Data Center and AI Technology Premiere,” AMD (NASDAQ: AMD) announced a continuation of its relationship with Amazon Web Services (AWS) with a preview of the next generation Amazon Elastic Compute Cloud (Amazon EC2) M7a instances, powered by 4th Gen AMD EPYC™ processors.

Key Points: 
  • SANTA CLARA, Calif., June 13, 2023 (GLOBE NEWSWIRE) -- Today, during the “ Data Center and AI Technology Premiere ,” AMD (NASDAQ: AMD) announced a continuation of its relationship with Amazon Web Services (AWS) with a preview of the next generation Amazon Elastic Compute Cloud (Amazon EC2) M7a instances , powered by 4th Gen AMD EPYC™ processors.
  • “AWS has worked with AMD since 2018 to offer Amazon EC2 instances to customers.
  • “When we combine the performance of 4th Gen AMD EPYC processors with the AWS Nitro System, we’re advancing cloud technology for our customers by allowing them to do more with better performance on even more Amazon EC2 instances.”
    “AMD and AWS are reimagining what’s possible with cloud performance, driving a differentiated offering for customers with the next generation of Amazon EC2 instances,” said Dan McNamara, senior vice president, general manager, EPYC Business, AMD.
  • “We continue to showcase the immense capabilities of the world’s best data center CPU, the 4th Gen AMD EPYC processors, and our ongoing collaboration with AWS highlights the momentum and demand for EPYC to power faster applications enabling customers to bring a broader range of workloads to the cloud.”
    The new Amazon EC2 M7a instances, using 4th Gen AMD EPYC processors are now available in preview.

Myrtle.ai Achieves 5.1 Microsecond Latency in Financial LSTM Inference Benchmark with VOLLO

Retrieved on: 
Thursday, June 1, 2023
STAC, Benchmark, Cloud, Heart, ML, PCI Express, LSTM, Half-precision floating-point format, Machine learning, SLM Solutions Group AG, Ecosystem, AI, Family, FPGA, Computer data storage, Cryptocurrency, Intel

[1] This is the first FPGA-based solution with published results for the Tacana Suite of STAC-ML™ and it achieved incredible latency results.

Key Points: 
  • [1] This is the first FPGA-based solution with published results for the Tacana Suite of STAC-ML™ and it achieved incredible latency results.
  • The STAC-ML Markets (Inference) benchmark represents the needs of capital markets firms using machine learning inference to respond rapidly to changes in the markets.
  • VOLLO achieved latencies as low as 5.08 microseconds with a throughput over 800k inferences/second.
  • For AI applications requiring low latency, like those in the STAC-ML benchmark, use of hardened bfloat16 decreases latency and increases throughput.

Myrtle.ai Achieves 5.1 Microsecond Latency in Financial LSTM Inference Benchmark with VOLLO

Retrieved on: 
Thursday, June 1, 2023
STAC, Benchmark, Cloud, Heart, ML, PCI Express, LSTM, Half-precision floating-point format, Machine learning, SLM Solutions Group AG, Ecosystem, AI, Family, FPGA, Computer data storage, Cryptocurrency, Intel

[1] This is the first FPGA-based solution with published results for the Tacana Suite of STAC-ML™ and it achieved incredible latency results.

Key Points: 
  • [1] This is the first FPGA-based solution with published results for the Tacana Suite of STAC-ML™ and it achieved incredible latency results.
  • The STAC-ML Markets (Inference) benchmark represents the needs of capital markets firms using machine learning inference to respond rapidly to changes in the markets.
  • VOLLO achieved latencies as low as 5.08 microseconds with a throughput over 800k inferences/second.
  • For AI applications requiring low latency, like those in the STAC-ML benchmark, use of hardened bfloat16 decreases latency and increases throughput.