Writing this blog feels like a version of The Odd Couple. Here we are at WEKA blazing trails and attacking the problems associated with the latest High-Performance-Computing, AI/ML and GPU-based Workloads. Oracle on the other hand is an enterprise database management company that powers the financials and business operations of tens of thousands of companies. Oracle Cloud Infrastructure (OCI) is a top-tier Hyperscale cloud that provides XaaS compute and application services to Oracle customers, and offers a lot more than you might expect, including AI/ML and GPU workloads.

Find out more info on HPC in OCI workloads here.

So how are Oracle and WEKA working together?

WEKA is proud to announce the availability of WEKA on Oracle Cloud Infrastructure.  Both companies have been working together to validate WEKA can run on OCI Bare metal instances to deliver the highest level of performance that customers demand for their critical workloads. In fact, the outcome has been extremely good with initial results showing the ability to provide nearly 2 Terabytes/sec of throughput and over 11 million IOPs of performance. Yes, you heard that correct: *Terabytes* per second! For more details on how this was achieved, check out the blog from Oracle here.

As OCI expands its portfolio of services, The WEKA Data Platform provides significant benefits to a variety of workloads on OCI including EDA, AI/ML, Life Sciences, and Financial Analysis. For performance, WEKA not only prevents “data stall” in your data pipelines by eliminating data copies in the process, but it also keeps GPU utilization rates high by providing massively performant IO to the applications regardless of whether it’s bandwidth or IOP driven. Traditionally, for each step in these workloads’ pipeline, you would need to copy data to an optimized location in order to quickly ingest data, then perform some type of ETL function, then do analysis/training/inference, and then send the data off to a data lake for archive or other analysis. WEKA accelerates this process by providing performance that works across both large and small IO concurrently. Zero copying is needed for each data pipeline which improves time-to-value of the data. An example of this for AI/ML is below:

Apart from performance, WEKA and OCI have also validated transparent storage tiering for WEKA by using S3 compatible OCI Object Storage as a low-cost tier. This enables customers to run at Petabyte scale in a high-performance system by leveraging local NVMe SSDs for hot data and object storage for warm/cold data.

OCI brings a strong set of capabilities into the cloud space. WEKA enhances that offering by accelerating the data pipeline used to feed those capabilities. The Odd couple? Maybe, but it sure is working well together.

For more information, please contact info@weka.io