Access your Pro+ Content below.
Disaggregated systems try to unlock the NVMe conundrum
This article is part of the Computer Weekly issue of 17 April 2018
The shared storage array has seen amazing success as a key component of IT infrastructure over the past 30 years. Consolidation of storage from many servers into one appliance has provided the ability to deliver more efficient services, increase availability and reduce costs. But as storage media moves towards the use of flash NVMe, shared arrays are showing their age and are being superseded by a new wave of disaggregated storage products. To understand the root cause of the impending issue with shared storage, we need to look at the media in use. Recently, hard drives have given way to flash (Nand) storage that is many orders of magnitude faster than spinning media, but that wasn’t the case for many years. The performance profile of a single hard drive was such that centralising input/output (I/O) through one or more controllers didn’t impact on performance and, in most cases, improved it. In other words, spinning disk drive hardware actually was the bottleneck in I/O. The controller provided much-needed functionality with no ...
Access this CW+ Content for Free!
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
Features in this issue
NVMe can unlock the true potential of flash, but storage controllers put a bottleneck back in the I/O path. We look at how suppliers are trying to solve the problem
Flexibility and low cost make the cloud well-suited to disaster recovery, but there is no one-size-fits-all route to cloud disaster recovery. We look at the key choices
Companies can spend millions, but half of SAP projects fail because there is no connection between the implementation and the original business case