Linux recovery history

Ten years ago, IBM built upon its leadership position in the storage hardware industry by becoming one of the first major vendors to develop a storage software product that automatically backed up files over the network to disk or tape.

The product, originally called IB MADSTAR Distributed Storage Manager (ADSM) launched on July 29, 1993, to little fanfare. Neither analysts, customers, nor even the core group of engineers who developed the product knew that this software solution would some day pave the way for a new era of storage management where engineers foresaw the need to integrate the backup and recovery cycle within an enterprise's overall systems management solution. This innovative and simple strategy would eventually mold the storage software market into the $5 billion industry it is today.

The Way It Was

In early 1993, there were only fifty Web servers operating worldwide. With the explosion of the World Wide Web, that number increased more than 30-fold in the course of 18 months, continuing to accelerate today at a pace that is virtually impossible to track. Today the power of Internet technologies, combined with the proliferation of communication devices, allows people to connect with virtually anyone, anywhere, anytime--with little or no cost. An idea that seemed incomprehensible to most of us ten years ago.

To get a feel for the storage industry that existed in the early 1990s, one must consider that the Storage Area Network (SAN) had not yet been invented. Most enterprises relied on mainframes while, today, networks built on scalable Intel-based servers running Windows, Unix or Linux are much more prevalent. Enterprises were just starting to come to grips with the fact that information important to their businesses was no longer stored centrally, but distributed to systems that were not under central control. Each year over the past decade, the average enterprise has doubled the amount of storage it requires, and according to IT analyst firm Gartner, backup software was the most commonly purchased enterprise software last year.

The Challenge

As mainframe computer systems were making the transition from colossal machines to more personalized and controllable systems, IT managers lacked the capability to manage the cumbersome environments. In these "antiquated" times, system administrators would set aside time during off hours and the weekends to run backup programs. In the newer networked world, IT managers needed to significantly shorten the backup window as data requiring backup began to multiply faster than ever.

IBM ADSM was developed by extending technology in a product that was originally named Workstation Data Save Facility (WDSF). WDSF started out as a utility at IBM's Almaden research center to backup distributed hosts to VM (a mainframe operating system called Virtual Machine). IBM's VM development team in Endicott picked up the utility and, together with the Almaden Research Center, turned WDSF into a stand-alone product.

At the same time, the storage product division at IBM was trying to develop a policy-based backup product for distributed systems. After some alternative efforts and evaluation of WDSF, the team approached the Almaden developers to build the distributed backup product that became IBM ADSM.

The first version of ADSM provided the backup function for distributed systems to MVS (Multiple Virtual Storage) and VM mainframe servers. Microsoft Windows, IBM OS/2, Apple Macintosh, Novell Netware and AIX clients were supported for backup/restore and archive/retrieve processing. The second version added AIX and OS/2 servers and associated tape devices together with client HSM functions on AIX. A client application programming interface (API) was introduced for use by developers and business partners in backing up databases or application servers (such as mail servers) whose requirements went beyond simply backing up the underlying file system.

After the acquisition of Tivoli Systems, IBM moved ADSM to the new systems management software group. This group continued to enhance many of the functions that define the product today including disaster recovery, hierarchical storage management, enterprise configuration, Web interfaces and exploitation of storage area networks. Finally, IBM ADSM became IBM Tivoli Storage Manager.

The Solution

New storage management software, led by products such as ADSM, ushered in a different kind of storage management that saw the integration of backup and recovery functionality with an enterprise's overall systems management solution. By providing a consolidated management console with broad platform support, these new management solutions were able to provide storage administrators with a simplified, easy to manage backup cycle.

Unlike competitors' products, IBM ADSM was developed with a flexible architecture that evolved from a mainframe-influenced, policy-based backup product into a broader storage management solution that now includes disaster recovery, space management, online database and application protection, sophisticated archive retention, and operational reporting.

Tivoli Storage Manager pioneered an architecture that used an embedded database to take backup functionality to a new level. Other vendors in the storage software space developed their own backup and recovery products primarily by enhancing stand-alone workstation backup tools. The competitive landscape has driven ongoing innovations that have anticipated the rapid changes of the IT industry and preemptively identified potential challenges before they emerged as problems. Such innovations as disk- to -disk backup, LAN-free backup, bare machine recovery and library sharing addressed customer pain points years ago as the industry was still evolving.

Almost from its inception, IBM Tivoli Storage Manager was an open product, exposing the client/server API to business partners to enable customers to store any data directly to the server. In addition, its flexibility and policy configuration have made it resilient to changes in both storage technology and customer use.

IBM Tivoli Storage Manager provides centralized storage management and data protection to a variety of industries including utility and power, entertainment, financial services, government, education, healthcare and the military. While primarily used as a backup/recovery product, its extensive support for archive packages, descriptions and retention policy has led to its widespread use for archiving information for legal or historical reasons. Some customers have even developed their own custom applications that store data directly into Tivoli Storage Manager through the API for both archive as well as backup purposes.

Why IBM Tivoli Storage Manager has Remained on Top

Today, IBM Tivoli Storage Manager protects millions of systems and applications around the world. It is the product of choice for more than 80% percent of Fortune Global 100 companies and 96 of the top 100 financial institutions. Tivoli and IBM combine to hold more than 224 patents related to storage management technology.

IBM Tivoli Storage Manager has remained a recognized market leader because it has stayed one step ahead of industry trends and developed sophisticated storage management technology to meet customer needs. IBM Tivoli Storage Manager supports a wide range of heterogeneous client and server platforms, providing it with the ability to offer unmatched protection and flexibility in managing growth.

Supporting more than 30 client platforms, eight server platforms and hundreds of I/O devices, IBM Tivoli Storage Manager provides centralized data protection, archiving, and HSM to heterogeneous environments. In addition, IBM Tivoli Storage Manager provides access to a range of high-performance peripherals, allowing the flexibility to utilize available media resources. It is unique in its ability to migrate managed data from one type of media to another. This hierarchical storage capability makes multiple systems backing up at the same time using disk-to-disk backups a reality. By backing up directly to disk, there is less need to use the technique of tape multiplexing. Multiplexing causes slow restores while the organized migration of data from disk to tape expedites restore times. Hierarchical server storage organization also assures the longevity of the data protected by Tivoli Storage Manager because it can automatically migrate to new media technology.

IBM Tivoli Storage Manager's embedded database architecture offers other advantages when compared to the traditional flat-file oriented storage products. For example, it features progressive backup which backs up only new and changed files, and which has no dependencies on full backups. This advanced technique reduces the amount of data being managed, and thus the time and bandwidth it takes to transfer the data and the amount of tape or disk resources needed to store the data.