Today, Hewlett Packard Enterprise (HPE) introduced new high-performance computing solutions that aim to accelerate HPC adoption by enabling faster time-to-value and increased competitive differentiation through better parallel processing performance, reduced complexity and deployment time.
These innovations include:
- HPE Core HPC Software Stack with HPE Insight Cluster Management Utility v8.0: Designed to meet the needs of businesses in any field, HPE is introducing a new highly flexible, simple and comprehensive software-defined platform for HPC that accelerates the configuration, deployment and management of clusters for HPC.
- System architecture innovations: To accelerate the performance of customers’ HPC workloads and reduce the complexity of their infrastructure, HPE is introducing systems innovations that build on the latest technologies from the HPE Apollo technology partner ecosystem.
- HPE ANSYS solution for CAE: To help customers scale their compute power, this solution is uniquely designed to handle large data sets and address the design prototyping needs of both large enterprises and mid-market customers and delivers improved application performance, faster resource provisioning and lower total cost of ownership.
HPC Systems Design Innovation (Apollo 2000 and Apollo 6000)
HPE is further enhancing the software defined Apollo HPC platform by introducing systems design innovations for the Apollo 6000 system with new HPE ProLiant XL260a server trays based on the next generation of the Intel Xeon Phi processor family and the Intel Omni-Path Architecture (Intel OPA) to reduce latency, and increase bandwidth and performance. HPE is also enhancing the Apollo 2000 system and ProLiant DL server platforms with Intel OPA fabric options. By combining these latest advancements in Intel Scalable System Framework with the scalability, flexibility and manageability of the HPE Apollo portfolio, customers will gain new levels of performance, efficiency and reliability. In addition, customers will be able to run HPC applications in a massively parallel manner with minimal code modification.
Bridges uniquely combines HPE’s purpose-built compute platforms optimized for our compute and data-intensive HPC environment with the Intel Omni-Path interconnect fabric designed to minimize latency,” said, Nick Nystrom, Principal Investigator for Bridges at the Pittsburgh Supercomputing Center (PSC). “The tight partnership and broad integration of technologies between Intel and HPE has been instrumental in enabling PSC to design Bridges and to place it into operation with a uniquely integrated infrastructure that delivers unmatched performance and flexibility.”
To provide customers with additional choice and flexibility, HPE is also enhancing HPE Apollo 6000 and HPE Apollo 2000 systems with next-generation EDR 100Gb/s InfiniBand solutions from Mellanox Technologies. Ideal for HPC clusters that require low latency and high bandwidth networking, Mellanox EDR fabric technology provides customers the network performance needed to improve response times and alleviate bottlenecks that impact application performance.