An IBM File System That Rivals Could Love

A few years ago, IBM promised a file system with multi-vendor interoperability.

On Tuesday, Big Blue made good on its word with the next iteration of its TotalStorage SAN File System (SFS). It supports storage devices from rivals EMC , HP , and Hitachi .

SFS, formerly code-named “Storage Tank,” was first launched last October as a heterogeneous file system that provides a single point to manage files and databases.

The software is designed to help reduce the amount of storage required in a customer’s infrastructure. It works by allowing storage resources to be shared more efficiently across servers via virtualization , which allows users to run multiple versions of an application or operating system on one physical device.

But the first version supported only IBM storage area network (SAN) devices. Analysts welcomed the news, but competitors jeered over its lack of interoperability.

As the result of work at its Almaden Research Center, IBM says version 2.1 of its latest SFS supports non-IBM hardware. The shift reflects the general trend of offering customers more choice by providing interoperability with rival vendors’ products. After all, customers are attracted to vendors that offer the greatest amount of choice, which saves them from the dreaded prospect of vendor lock-in.

Last month, IBM announced its TotalStorage SAN Volume Controller (SVC), which allows users to virtualize storage to manage multiple devices and free up capacity in their storage SANs, can now pool data from EMC machines.

SAN File System Version 2.1, which also features improvements to the administrative user interface and support for Red Hat Linux Enterprise Server 3.0 and Sun Solaris 9, will be available June 29. The updated SFS Metadata Server will cost $10,000 per processor plus a usage charge of $5,000 per processor for connected application servers.

Big Blue’s SFS 2.1 news also comes two weeks after the company held a media event in Cambridge, Mass. to announce its pledge of support for information lifecycle management (ILM) as a key
component of its on-demand strategy.

At that event, Jai Menon, IBM Fellow and director and chief technologist of storage systems and design, detailed a number of forward-looking research
products from the company’s Almaden Lab that highlight autonomic computing and virtualization.

Menon told the company’s Intelligent Storage Bricks (code-named Ice Cube) project is taking the supercomputing approach and applying it to a hybrid server/storage system, which is essentially a square block of hardware made up of several cubes.

The cubes, independent computing servers or storage servers that perform virtualization tasks independently, essentially make up an entire data center. The machine is perhaps three years or more from seeing the light of day and is intended as the company’s replacement for its TotalStorage Shark
Enterprise Storage Servers, according to Menon.

Story courtesy of

Back to Enterprise Storage Forum

Clint Boulton
Clint Boulton is an Enterprise Storage Forum contributor and a senior writer for covering IT leadership, the CIO role, and digital transformation.

Latest Articles

Top Managed Service Providers (MSPs) for 2021

Managed service providers perform an arsenal of outsourced tasks, including cloud deployment and migration, advanced data analytics, and IoT and network installation. Managed services...

7 Essential Compliance Regulations for Data Storage Systems

Many of the compliance systems that companies have to deal with require the same kind of actions, processes, and plans. Here are key elements.

DNA Data Storage: Could Data Files Be Stored as DNA?

Using DNA storage for digital data is a well established idea. Here’s how it could come to full fruition as acceptance widens.