Deliver the performance required to power your most demanding applications to dramatically accelerate time to insights.
Weka’s modern architecture is built from ground up for flash and optimized for NVMe and cloud to power breakthrough innovation.
HOW FAST IS WEKAFS
BY INDEPENDENT ORGANIZATIONS
BY INDEPENDENT ORGANIZATIONSSPEC SFS, IO-500 and the STAC Benchmark Council have all ranked WekaIO as a top performer . In the June 2020 benchmarks WekaIO set 12 STAC-M3 world records for mean query-response times and 5 world records for throughput (more here). In the WekaIO 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. (more here).
BY REFERENECE ARCHITECTURE
BY REFERENECE ARCHITECTUREThe NVIDIA and Weka AI DGX POD™ reference architecture (RA) powered by NVIDIA DGX™ A100 and Weka’s industry-leading file system WekaFS demonstrated 162GB/sec of throughput as well as 970K IOPs leveraging NVIDIA® GPUDirect® Storage (GDS).
Download the reference architecture here.
BY CUSTOMERSMicrosoft Research using NVIDIA DGX-2 server with dual-port NICs non-standard configuration achieved record performance using 16 NVIDIA V100 GPUs with NVIDIA® GPUDirect® Storage (GDS).
Click here to learn more.
Our mission is to make your storage radically simple. Our simplicity is evident whether you look at the user interface or try some tasks that are difficult if not impossible in legacy architectures.
Weka delivers the performance required to power your most demanding applications. Now what would you do if you had this kind of performance at your disposal?
Scale your compute and storage independently and linearly on-premises or in the cloud without limits.
WEKA BREAKS RECORDS ON INDUSTRY STANDARD BENCHMARKS
The Aiden Lab cluster required a new solution to improve application performance and facilitate the deployment of a high-performance file system in a cloud computing environment. We required a solution that could support the team and their research related to Genome Architecture and felt that neither GPFS nor Lustre could keep up with our workload.David Weisz, Lead Scientific Programmer, Aiden Lab At Baylor College Of MedicineRead More