Get storage off expensive tier 1 ASAP, says Tarmin as it launches Gridbank 1.5

Tarmin
Technologies has announced version 1.5 of its GridBank object-based storage
archiving solution—with 64-bit Linux and Windows Server 2008 editions and new
features especially focused on Microsoft’s SharePoint.

The UK-based start-up
says its aim is to greatly shrink an organisation’s storage and data management
total cost of ownership (TCO) to be achieved not least by automatically
shifting data away from expensive tier 1 storage at the earliest opportunity.

GridBank
is an early example of a new breed of data storage solutions designed to automate
data management through applying policies—and it is these policies that are
used to quickly shift data away from expensive real-time tier 1 storage on to
tier 2 and archive.

Its
enterprise-class active archiving is fairly heterogeneous—supporting Windows,
Linux and several UNIX flavours, as well as VMware and HyperV virtual
environments. So a user’s CAPEX and OPEX can be
reduced partly by it making use of existing industry-standard server and
storage platforms in forming its grid-based active archive whose scalability is
to petabytes.

Eric Herzog, Tarmin’s
VP of marketing and sales, told me the company’s assessment of high end tier 1
15K rpm disk storage pricing versus tier 2 (e.g. SATA) typically showed a
difference of 3–4 times—but could reach even 7+ times—for the same quantity
of data with the same security. Moreover, while SNIA conservatively estimated
80% of data did not change after 90 days a survey of 900 mid-sized companies in
Europe and US indicated at least 60% of the
data did not change after five minutes of life!

So herein is a
clear general message: Get the data off tier 1 as soon as possible as it will
dramatically cut equipment CAPEX and OPEX without negatively impacting performance.
Gearing policies to automate this process is a key part of what GridBank is
about.

However, Herzog
was anxious to point out that there was much more to this CAPEX/OPEX saving:
“We manage virtual pools and don’t care which hardware vendors—and thin
provisioning means a ‘pay as you grow’ model.” In fact that is the way the
software is priced; user licensing is for the whole product, the user paying
only according to the amount of data being protected by the software.

Connection is IP
(LAN or WAN) allowing remote archiving with policies also including
identity-based security with support for LDAP and Active Directory (AD) at
present. Metadata indexing can also be customised for more fine-grained
migration of data.

At the other end
of the data’s life, GridBank supports the US Department of Defense (DoD)
standard for digital data shredding; by automating this through policies the
user can avoid forgetting to do this to comply with regulations leaving it open
to litigation. Also pretty tight is support for nine levels of encryption,
ranging from 448 bit Blowfish down to 64bit encryption or none. Everything is
set during installation simply by radio-buttons; applying policies is a
decision-tree process.

Version 1.5 includes
seamless integration with SharePoint. GridBank supports Microsoft Office SharePoint Server (MOSS) 2007 and Windows SharePoint
Services (WSS) 3.0; data storage policies can be applied to SharePoint data.
There is now also support for over 500 different file formats which
considerably broadens GridBank’s indexing, search and e-discovery capabilities.

The
web-based management console adds flexibility to management while custom policy
template creation can help speed deployment when, say, new server resources are
brought on-line. The metadata for policy action is limited to such things as file
type, usage analysis, capacity consumption and file ageing; it does not delve
into file content.

Also
new is support for the Representational State Transfer (REST) industry-standard
protocol to allow other applications to communicate with GridBank; it supports
NTFS, CIFS, HTTP, WebDAV, FTP and a number of other industry-standard
protocols.

What the software
does not support is structured databases (other than indirectly SQL Server
through SharePoint’s use of it). Here Herzog pointed out the software should
reduce the cost of SQL licences because of freed-up space and reduced data. IT
supports single instancing (SI) of files and versioning (but not, for instance,
de-duplication or CDP).

A secure,
policy-based data management approach is likely to is become ever more attractive
because it makes for a comparatively straightforward way of complying with
changing regulations and corporate governance requirements for data. A search
engine can assist in data discovery requests through retrieval when litigation kicks
in but the quality and scope of the search engine needs consideration. (For
instance, Herzog said GridBank’s search engine can be set to access data sources
otherwise outside GridBank protection.)

The company formally
launched its GridBank ILM “cradle to grave” data storage management solution
late last year – and general availability was as recent as February this year.