WekaIO introduced a transformative cloud-native storage answer underpinned by the world’s quickest file system, WekaFS, that unifies and simplifies the information pipeline for performance-intensive workloads and accelerated DataOps.
Weka has developed reference architectures (RAs) with main object storage know-how suppliers, like Amazon Internet Companies (AWS), Cloudian, IBM, Seagate, Quantum, Scality, and others in Weka’s Know-how Alliance Program, to ship cost-efficient, cloud-native knowledge storage options at any scale.
And Weka’s OEM partnership with Hitachi Vantara will ship an built-in end-to-end stack answer primarily based on the Hitachi Content material Platform.
WekaFS supplies the benefit of managing petabytes of information in a single, unified namespace wherever within the pipeline the information is saved, whereas additionally delivering the very best efficiency to speed up synthetic intelligence/machine studying (AI/ML), genomics analysis, high-performance computing (HPC), and high-performance knowledge analytics (HPDA) workflows.
Weka’s unified storage options with cloud-native ecosystem companions present the next buyer advantages:
- Quicker actionable enterprise intelligence from a single high-performance storage answer
- Value-efficiency with the flexibility to handle, scale, and share knowledge units
- Operational agility eliminating storage silos throughout edge, core, and cloud
- Enterprise robustness and safe knowledge governance
Handle extra petabytes of information cost-effectively and with fewer sources
Extending the WekaFS namespace from high-performance flash to an Amazon Easy Storage Service (S3) REST-enabled cloud object storage system is a less complicated and extra cost-efficient technique for managing petascale datasets with out compromising efficiency.
The filesystem metadata resides on flash whereas seamlessly extending capability over object storage, non-public or public. All of the I/Os are serviced by the flash tier whereas leveraging the item tier for capability scaling.
WekaFS permits knowledge portability throughout a number of consumption fashions supporting each non-public and public clouds with the flexibility to increase the namespace throughout each. A cloud-first mannequin delivers the very best storage effectivity and TCO throughout consumption fashions and knowledge tiers.
Facilitating knowledge safety, mobility, and DR
As knowledge has turn out to be a strategic asset for companies, lifecycle administration is paramount. Nonetheless, the datasets encountered in AI/ML, genomics, HPC, and HPDA have grown so large and agile that conventional backup and DR purposes fall quick, creating siloed namespaces and workflows which might be missing operational agility and knowledge safety.
Knowledge versioning is achieved utilizing Weka’s prompt and space-efficient snapshots functionality for experiment reproducibility and explainability. The snap-to-object characteristic captures a point-in-time copy of your entire, unified (flash and object retailer) file namespace that may be offered as one other file namespace occasion in a non-public or public cloud.
Weka’s built-in snapshots and end-to-end encryption options guarantee knowledge is at all times backed up and safe all through its lifecycle. WekaFS additionally supplies immutability and knowledge mobility for these datasets with prompt restoration.
Weka has partnered with main non-public and public cloud companions to make sure a completely validated and performant storage answer ecosystem, together with these licensed options: AWS S3, AWS Outposts, Cloudian HyperStore, Hitachi Content material Platform (HCP), IBM Cloud Object System (IBM COS), Quantum ActiveScale, and Scality RING.