Sign up for our newsletter and get the latest big data news and analysis.
Daily
Weekly

Weka and HPE Launch Joint Offering for Containerized Workloads

CAMPBELL, Calif., December 7, 2021 –AI data platform company WekaIO (Weka) today announced it has developed a solution with Hewlett Packard Enterprise (HPE) for demanding AI and ML workloads, now available on the HPE Ezmeral Marketplace.

HPE Ezmeral Runtime Enterprise is the industry’s first Kubernetes platform supporting both cloud native and non-cloud native applications. HPE Ezmeral supports hybrid deployments with both on-premises and public cloud support at the edge, core and cloud for a broad range of use cases like autonomous vehicles, single instance databases and CI/CD pipelines. Weka, HPE Ezmeral Runtime Enterprise and HPE Ezmeral MLOps enable customers to overcome the challenges of Kubernetes and AI/ML data pipelines and deliver value from the data center to the cloud to the edge (aggregation) with at-scale performance.

As part of this latest integration, Weka joins HPE’s growing ecosystem of partners that helps bring speed and agility to the application development lifecycle. The HPE Ezmeral Marketplace leverages the open-source Kubernetes ecosystem and its modern ISV partners to provide faster time to value. Together, they make it easier for clients to adopt integrated solutions that combine HPE Ezmeral software with validated industry-leading, third-party, commercial, and open-source applications. Weka can be found on the HPE Ezmeral Marketplace at https://www.hpe.com/us/en/software/marketplace/weka.html

Weka offers a high-performance data platform that was built to address the challenges posed by modern applications that leverage cloud, GPU compute through innovations in how Flash is used. Built on WekaFS, it offers the simplicity of NAS, the performance of SAN or DAS and the scalability of object storage. No more compromises between simplicity, speed or scale. With its unique software-defined architecture, customers can run on-premises, natively in the cloud, and orchestrate their data effortlessly between the two. The ability to mix and match Flash and Disk offers customers the best economics. This new Ezmeral validated Weka solution is in addition to the existing HPE ProLiant validated solutions currently available from HPE at HPE Solutions for Weka.

“Weka provides our customers with a high performance, software-defined storage platform to support the end-to-end AL/MLOPs data pipeline,” said Matt Hausmann, HPE Ezmeral Marketing & GTM. “HPE Ezmeral and Weka’s file storage platform offer customers the ability to build an application and deploy it anywhere with enterprise-class security, performance and reliability.”

While containers have become popular because of their lightweight nature, their immutability and portability, and their agility with orchestration provided by Kubernetes, persistent storage, data management and disaster recovery remain top barriers to adoption. Kubernetes at scale can be difficult. A validated solution like the HPE Ezmeral and Weka offering provides enterprises with benefits including faster time to market, operational agility, scale and ease of deployment.

“Weka sees a growing need to deploy stateful container applications on high-performance storage,” said Shailesh Manjrekar, Head of AI and Strategic Alliances at WekaIO. “Weka is excited about partnering with HPE to enable solutions with Weka’s scalable parallel file system on HPE Ezmeral Runtime Enterprise. This joint solution is optimized for NVMe and the hybrid cloud, targeting the most-demanding storage, AI/ML and data-intensive technical computing workloads.”

Weka’s software-based, cloud-native data platform allows customers to leverage low-cost S3 storage with fast NVMe flash storage, on-premises or in the public cloud, to quickly build out a massively scalable foundation for Enterprise AI projects that attain industry-leading performance and economics. Weka’s extensively patented zero-copy architecture, driven by high-performance, scalable multi-protocol support for all file and object, means that customers don’t have to buy, manage, and administer several distinct storage systems across ingest, processing and data protection.  Instead, they can leverage Weka to seamlessly integrate flash, hard drives and cloud S3 storage in a single data pool that scales to hundreds of petabytes in a single namespace.

To learn more about the Weka/HPE partnership and its joint value proposition in containerized environments, interested parties can visit the HPE blog at https://community.hpe.com/t5/HPE-Ezmeral-Uncut/High-performance-Kubernetes-workloads-with-HPE-and-Weka/ba-p/7155633

 

Leave a Comment

*

Resource Links: