HPC and AI – The bandwidth bottleneck explored

Architecture Guide for HPE Servers and WekaIO Matrix

Cover

For high-performance computing (HPC) and deep learning clusters, the improvements in compute speed by CPUs and GPUs pose a problem: compute speed has dramatically outpaced bandwidth improvements in storage solutions historically used by these applications.

Even if app capacity demands aren’t high, a significant number of storage devices are still needed to hit high bandwidth requirements—particularly with hard drive-based solutions.

This guide presents an alternative architecture—a highly scalable software-defined storage system designed to meet the latency and IOPS performance requirements of these HPC workloads.

For an architectural overview, use case examples, and the cluster design, read on here.

Vendor:
WekaIO
Posted:
03 Apr 2019
Published:
31 Dec 2018
Format:
PDF
Length:
19 Page(s)
Type:
Resource
Language:
English
Already a Bitpipe member? Login here

Download this Resource!