Network Computing is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Cloud Update: Suppliers

Cloud storage suppliers are the arms merchants for cloud storage. These
are the guys that provide the equipment that cloud storage providers use
to store user data. This layer of cloud storage, similar to Cloud
Providers

and Cloud ISVs,
has also matured quite a bit in the last year.

First, there is a better understanding of what it takes to be a cloud
storage supplier. Just because a vendor makes a RAID array does not mean they
have a cloud storage offering. At their core, systems that cloud storage
suppliers provide need to be cost effective and massively scalable.
There are traditional systems that meet that requirement, of course, but
that does not make them cloud storage. What does is when these systems
are coupled with cloud storage software that delivers the multi-tenant
cloud functions that providers need to manage multiple users and
organizations. In most cases, this also
includes a REST API set that allows the ISVs to control specifically how
data is transferred and stored on the system. Probably the biggest
development on the supplier side is the availability of cloud enablement
software that
provide cloud-like services that traditional manufacturers can add to their systems.

Most systems today offer some sort of scale-out capability where
performance and capacity can be incrementally increased by adding
another storage component or node. There is an interesting debate around
how these nodes should intercommunicate. A global file system, for
example, will allow multiple NAS heads to be treated as one from a file
storage perspective but it will not, typically, allow for all those NAS
heads to be administered as one. Global file systems do provide
more flexibility around node selection. A more grid-like architecture
does not provide as much flexibility in node selection but does provide a
single file system and a single point of management. Other systems have
added location independence to their scale out capabilities. In other
words, all the nodes of a cluster do not need to be in the same physical
location.

There are other capabilities that may or may not be important to the
provider and users of this storage infrastructure. One is global
positioning of data. This is more than just replicating data for a
disaster recovery standpoint. The goal is to move data as close to the
consumers of that data as possible. Another capability that ties into
this is the ability to make multiple copies of a popular dataset on
multiple nodes geographically dispersed around the world or country.

Cloud storage suppliers are often thought of as NAS only systems, where
performance plays second fiddle to capacity and scale. Performance
should not be overlooked, especially in cloud compute
environments. In these environments, the users are remote, but the data is
not. Companies providing cloud compute have the biggest challenge in
that not only do they need capacity and scale, they also need performance with the same competitive price as cloud
storage providers. Even though most users never see it or even know who made it, the actual
hardware and software that cloud providers use does make a
difference and does impact them directly. The selection of the right
hardware and software allows the provider to keep prices low and service
high. Most importantly it can keep them in business.