WekaIO Matrix - Storage for AI and Deep Learning

WekaIO - Overview

  • scale out storage cluster for maximum GPU utilization
  • for AI and deep learning applications with very high data throughput
  • over 10 GB/s to a single GPU client
  • four nodes in one enclosure (min. 8 nodes required)
  • parallel and distributed file system (for data and metadata)
  • virtually unlimited capacity scaling
  • automatic tiering to S3 object storage

WekaIO - fully customized for your business

We will be pleased to send you a non-binding offer by e-mail - for inquiries on working days until 4 pm on the same day!

You have questions about the configuration?

Give us a call:
+49 (0) 711 - 70 70 91 70

Product information

Modern analytics platforms are GPU intensive and require large data sets to deliver the highest levels of accuracy to the training or analytics systems. They also require a high bandwidth, low latency storage infrastructure to ensure a GPU cluster is fully saturated with as much data as the application needs.
Typical data sets can span from terabytes to tens of petabytes, and the data access pattern for each epoch is unique and unpredictable. This calls for a data infrastructure that can instantaneously and consistently feed large amounts of random data to multiple GPU nodes in real-time, all emanating from a single shared data pool.
WekaIO Matrix is the world’s fastest and most scalable file system for these data intensive applications, whether hosted on premises or in the public cloud. It has proven scalable performance of over 10GBytes per second bandwidth to a single GPU node, delivering 10x more data than NFS and 3x more than a local NVMe SSD.
EUROstor offers this solution in server enclosures, each containing four nodes. Two of these systems (a total of 8 nodes) are minimally required.

Model overview

WekaIO Storage Cluster
pdf offer
SPECIFICATIONS:
Min. Number of Nodes 8
Server standard server with four nodes on 2U, Intel architecture
Netzwerk-Interface 10, 25, 40, 50 or 100 Gbit Ethernet (several ports as needed)
Software WekaIO Matrix
further details cf. datasheets