Weka.io
Performance of NVME storage with scalability and economics of the cloud
WekaIO Matrix and Scality RING together provide the ideal storage solution for demanding AI/ML and HPC workloads.
The combined solution offers the performance of an NVMe™ array for processing hot data and the capacity of a limitlessly scalable, cost-effective private cloud storage for tiering and staging cold data in and out — all under a single namespace for easy management and optimized operational workflow.
Scality RING holds the master copy of warm buckets while indexers’ local storage is used for hot and cache copies of warm buckets. With most data residing on Scality RING, the indexer maintains a local cache that contains a minimal amount of data: hot buckets, and copies of warm buckets participating in active or recent searches, and bucket metadata. This renders indexers stateless for warm and cold data, boosting operational flexibility and agility.
Benefits
PERFORMANCE
Purpose-built for I/O-bound use cases, such as AI/ML and HPC workloads, the solution delivers the highest bandwidth and lowest latency, plus linear scaling. Remove storage bottlenecks and make workloads compute-bound again.
SCALABILITY & AVAILABILITY
Gain a limitless storage pool with industry-leading data durability and protection. Easily and seamlessly grow storage capacity on-demand to petabytes and beyond, while keeping all data sets online and always available.
COST-EFFICIENCY
Software-defined, hardware-independent storage runs on commodity servers, with no vendor lock-in or premium. Built-in resiliency eliminates the need for prohibitive data backup at scale.
SIMPLICITY
Get automatic, transparent data tiering and migration under a single namespace. Snapshot capabilities enable enhanced disaster resilience and cloud bursting.
The solution
Learn why the joint solution from WekaIO and Scality is perfectly suited to meet the demands of AI/ML and HPC workloads.
Webinar replays
ACCELERATING CRYO-EM & GENOMIC WORKFLOWS
Busting Myths with Scality & WekaIO
SUPERCHARGE YOUR HPC AND AI WORKFLOW IN FEDERAL