Fujitsu goes hyperscale with CD10000 open-source storage

In a tilt at the hyperscale market the CD10000 marries Fujitsu hardware with Red Hat Ceph open-source storage software

Fujitsu has announced the CD10000 storage platform, which has storage nodes that can scale to exabytes of capacity based on OpenStack open-source storage software from Red Hat's Inktank Ceph Enterprise (ICE).

The CD10000 platform (CD meaning content depot) is aimed at large-scale users such as service providers, telcos and media organisations that will store lots of data for a long time.

The IT landscape, to which CD10000 is addressed, is IDC’s so-called third platform – also cited in EMC’s ViPR release – in which organisations generate and retain data from multiple sources beyond the datacentre, such as machine-generated information, or mobile and bring-your-own-device (BYOD) data.

Fujitsu storage product development head Marcus Schneider said the product is also a response to the “serverisation” of storage and “the separation of storage hardware and software”, although the CD10000 comprises Red Hat storage software tied to Fujitsu software.

A CD10000 deployment can start at four nodes. Nodes come in three flavours: general purpose, with 12TB of serial-attached SCSI (SAS) capacity; performance, with 35TB including flash and SAS drives; and capacity, which come with 252TB of nearline-SAS drives.

Nodes can be mixed and matched to provide, for example, combinations of performance and capacity-oriented storage. In the initial releases, CD10000 will scale to 200 nodes.

Storage software is provided by Red Hat, which upgraded its OpenStack storage software distribution ICE to version 1.2 in July 2014.

ICE is Red Hat’s commercially-supported distribution of Ceph, an open-source product that currently supports only object storage and block access.

CD10000 connectivity is block (iSCSI) and object (Amazon S3) at present, although file access is in the roadmap, said Schneider.

The upgrade to version 1.2 saw the addition of erasure coding to provide cost-efficient backup and archiving, as well as tiered storage functionality.

Automated storage tiering in ICE allows data to be moved between tiers of storage defined by performance and data protection method.

For example, an image uploaded to a web page and subsequently unused would reside in the least costly, slowest-to-access drives, protected by erasure coding.

Read more on SAN, NAS, solid state, RAID