It appears you have not yet registered with our community. To register please click here...

 
Go Back [M] > Madshrimps > WebNews
Toshiba Announces KumoScale NVMe over Fabrics Software Toshiba Announces KumoScale NVMe over Fabrics Software
FAQ Members List Calendar Search Today's Posts Mark Forums Read


Toshiba Announces KumoScale NVMe over Fabrics Software
Reply
 
Thread Tools
Old 23rd March 2018, 11:43   #1
[M] Reviewer
 
Stefan Mileschin's Avatar
 
Join Date: May 2010
Location: Romania
Posts: 148,597
Stefan Mileschin Freshly Registered
Default Toshiba Announces KumoScale NVMe over Fabrics Software

Today, Toshiba is launching a new software solution for pooling NVMe-based storage in a datacenter environment. The new KumoScale software implements the NVMe over Fabrics protocol (NVMeoF) and allows storage to be access over the network with less than 20µs of added latency compared to NVMe SSDs directly attached to the local system. KumoScale provides abstraction over the individual SSDs so that multiple drives can form a shared pool of storage, from which separate volumes can be allocated to serve as storage for different services running on compute-oriented servers elsewhere in the datacenter.


Toshiba announced their NVMeoF work last year at Flash Memory Summit in August. Today, they are giving it a brand name, announcing a general availability release, and sharing more about how it works—both technologically and commercially. KumoScale is a pure software product. While it is intended to complement Toshiba's datacenter SSDs, it runs on existing commodity servers, can use any NVMe SSDs, and makes the storage available over the standard NVMeoF network protocol. Toshiba has partnered with Portworx to enable integration between KumoScale and popular cloud computing infrastructure management tools like the Kubernetes container orchestration system. The combined stack of Portworx software and KumoScale-managed storage is touted as allowing compute and storage resources to be completely disaggregated, preventing "stranded" compute power or storage performance that can result from direct-attached storage configurations.

Toshiba's internal testing of KumoScale has demonstrated scalability of storage nodes to over 8M random read IOPS, capacities of 384TB, and partitioning of storage into over 4000 separate NVMe namespaces. Toshiba and Portworx are demonstrating KumoScale this week at the Open Compute Project (OCP) Summit in San Jose.

https://www.anandtech.com/show/12532...brics-software
Stefan Mileschin is offline   Reply With Quote
Reply


Similar Threads
Thread Thread Starter Forum Replies Last Post
Kingston expands its HyperX Fury and Impact DDR4 memory lineup Stefan Mileschin WebNews 0 23rd March 2018 11:43
Kingston HyperX Fury 16GB DDR4-2666 Memory Kit Stefan Mileschin WebNews 0 12th April 2017 13:51
Kingston HyperX Fury 2666 MHz 16GB DDR4 Memory Kit Review Stefan Mileschin WebNews 0 16th October 2015 09:38
HyperX announces new additions to FURY and Impact DRAM lineup Stefan Mileschin WebNews 0 1st September 2015 11:50
HyperX Releases Additions to FURY, Impact Line of DRAM Stefan Mileschin WebNews 0 1st September 2015 11:42
Kingston Also Shows off HyperX Predator DDR4 and HyperX Fury DDR4 Memory Stefan Mileschin WebNews 0 13th January 2015 08:14
Kingston HyperX Fury Red DDR3 1866MHz 16GB Memory Review Stefan Mileschin WebNews 0 23rd June 2014 07:36
Kingston Serves Up HyperX Fury Series DDR3 Memory Stefan Mileschin WebNews 0 14th January 2014 06:23
Kingston Technology Expands HyperX Memory Family Stefan Mileschin WebNews 0 16th August 2012 10:29
Kingston Technology Expands HyperX Product Line with New Memory Cooling Fan jmke WebNews 0 5th June 2009 16:29

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


All times are GMT +1. The time now is 20:17.


Powered by vBulletin® - Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO