Rescale, the leading platform for high performance computing (HPC), is excited to publicly announce that Amazon EC2 Hpc7g instances are now available on Rescale under the coretype name “Pectolite.”
With the addition of Hpc7g, Rescale customers can easily access the power of Amazon Web Service’s (AWS) Arm-based Graviton3E chip architecture for powering their R&D innovation efforts. The new Pectolite coretype offers industry-leading cost-performance and energy efficiency for running both high performance computing and high-throughput computing to power cutting-edge digital R&D.
The nature of supercomputing for research and engineering has been undergoing a transformation driven by the continuing growth of the cloud. Because of the cloud’s economies of scale, leading service providers like AWS have been able to drive the development of specialized, high performance hardware infrastructures available on-demand through cloud data centers.
As one of the industry’s leaders, AWS has continually pushed the boundaries of innovation in cloud HPC technologies. Their work with Arm has been foundational to AWS’s pioneering effort.
AWS’s HPC chip architectures are purpose-built to offer exceptional price-performance for running HPC workloads at scale. AWS’s HPC cloud services are ideal for compute-intensive applications such as large, complex simulations and deep learning workloads.
“The ability to rapidly iterate and improve product design using digital twins and simulations is important to our customers, who also increasingly care about environmental sustainability, in addition to performance and flexibility,” says Mulyanto Poort, vice president of HPC engineering at Rescale. “The introduction of Hpc7g will help our customers further accelerate their digital engineering efforts while reducing their carbon footprint.”
Major Gains in Cost-Performance and Energy Efficiency
Hpc7g instances are powered by custom-designed chips from AWS and are engineered to deliver exceptional performance and efficiency for demanding R&D computational workloads.
“We are excited to offer Hpc7g instances. Their floating-point performance and expanded AWS Elastic Fabric Adaptor (EFA) bandwidth will be very beneficial for key HPC use cases,” Poort says. “For example, we’ve seen a three-fold increase in cost-performance compared to the previous generation of Graviton processors in mission-critical customer workloads like computational fluid dynamics (CFD).”
Hpc7g Technical Specifications
Hpc7g builds on the success of AWS Graviton-based cloud services with impressive technical capabilities for HPC applications. With these instances, you can carry out complex calculations across HPC clusters using up to tens of thousands of cores with high performance and lower costs.
Custom-Designed AWS Graviton3E Processors
AWS Graviton3E processors are at the heart of Hpc7g instances. AWS has designed these Arm-based chips to deliver outstanding performance for various HPC applications. With up to 64 vCPUs, Hpc7g offers exceptional parallel processing capabilities, allowing users to tackle complex computational tasks like high-throughput computing cost-effectively.
High Memory and Networking Performance
Hpc7g comes equipped with substantial memory and networking capabilities. Users can handle large-scale simulations, data analytics, and other memory-intensive workloads with up to 128 GiBs of memory per instance and up to 200 Gbps of network bandwidth.
Rescale Empowers R&D Professionals with Hpc7g
Rescale users can now access the exceptional performance and scalability of Hpc7g seamlessly through Rescale’s intuitive interface and workflow automation. Using Hpc7g on Rescale eliminates complex infrastructure and software management challenges, letting R&D professionals focus on their research and innovation. With Rescale, users can instantly spin up a Hpc7g instance, as well as having on-demand access to 1,000+ engineering, scientific, and AI applications that are pre-installed on Rescale.
The advanced capabilities of Hpc7g and Rescale’s intelligent platform open up new R&D innovation potential for how organizations take advantage of advanced digital engineering methods such as computational fluid dynamics, finite element analysis, molecular dynamics, genomics, bioinformatics, data analytics, and machine learning.
Thanks to the technical innovation of Hpc7g cloud services, researchers, engineers, and scientists across industries including aerospace, automotive, energy, life sciences, manufacturing, public sector, and semiconductors can accelerate their workloads, solve complex problems, and achieve unprecedented R&D breakthroughs with AWS and Rescale.