Beating out Massive Supercomputer Installations, Weka Delivers Substantial Performance Improvements for HPC, Leveraging Best-In-Class File System Technology on Amazon Web Services
WekaIO, an innovation leader in high-performance, scalable file storage for data-intensive applications, announced that it has been awarded the first-place ranking on the IO-500 Challenge, rankings that compare storage systems that work in tandem with the world’s largest supercomputers.
“This benchmark reinforces how Weka can deliver superior performance density in a more user-friendly manner when compared with the largest supercomputer storage systems.”
The Weka File System (WekaFS™) running on Amazon Web Services (AWS) set a record on the IO-500 benchmark with an overall number of 938.949—51% better performance than the prior record set at International Supercomputing Conference 2019. The overall performance record achieved demonstrates that high-performance computing (HPC) on the cloud can compete with the most advanced on-premises supercomputer clusters. WekaIO received the award for the highest overall number as well as the highest metadata performance, demonstrating that its distributed metadata is a game changer for higher performance computing.
Each system was evaluated by the IO-500 benchmark that measures the storage performance using read/write bandwidth for large files and read/write/listing performance for small files. Storage performance in supercomputers is a critical factor for large-scale simulation, big data analysis, and artificial intelligence (AI). The IO-500 list serves as a resource for companies looking to improve their storage performance, which in turn helps to improve processor performance and utilization.
“The results of the IO-500 benchmark validate our leadership position for storage solutions in the HPC market and illustrate the performance benefits associated with using WekaFS on AWS. Weka provides organizations with a way forward to solve tomorrow’s biggest computing problems. It’s is a modern file system architected to provide extreme performance at any scale,” said Liran Zvibel, CEO and co-founder, WekaIO. “This benchmark reinforces how Weka can deliver superior performance density in a more user-friendly manner when compared with the largest supercomputer storage systems.”
WekaFS is a shared, parallel file system that delivers the highest bandwidth, lowest latency performance to any InfiniBand- or Ethernet-enabled GPU- or CPU-based cluster. The POSIX-compliant file system leverages an innovative, highly-optimized software stack that leverages modern architectures such as low-latency NVMe-oF and massively distributed metadata. Built for enterprise-grade HPC, it tightly integrates object storage for best economics at scale and supports features such as snapshots, encryption, authentication, high availability, and cloud backup. Weka fully supports legacy protocols such as NFS and SMB to provide end-user access.