Information lifecycle management (ILM) requires tools for tasks such as information classification, data classification, and data movement in conjunction with a tiered storage architecture.
By Michael Peterson
Tiered storage is an important strategy for IT services. Yet among the many discussions of the value of tiered storage from a business perspective or its usefulness for IT in reducing costs, some vendors insist on selling tiered storage as information lifecycle management (ILM) as if this were all there is to ILM. Not so. Tiered storage is an important component of an ILM practice, but only one component among many.
Tiered storage establishes a hierarchy of storage systems based on service requirements and cost. Why leave inactive or expired data on your expensive primary storage arrays? The cost advantages of tiered storage can be immense. Most enterprise implementations see a return on investment in less than six months, with reductions in purchasing and operating costs in the millions of dollars per year, according to Strategic Research Corp.
Adoption of tiered storage in data centers is well underway, fueled in part by the explosion of low-cost, high-capacity Serial ATA (SATA) disk arrays. These capacity-oriented systems have found a home in tiered storage architectures for secondary storage, backup, online archives, compliance stores, and other applications.
To combine ILM-based practices with tiered storage, first identify the value and requirements (information classification), set policies and service level requirements that map to your storage resources (data classification), and then implement tiered storage levels.
By applying policies that define information, data, and security service requirements such as retention, placement, protection, business continuity, archive, and compliance policies to information based on its value or business requirements, ILM practices can be automated while reducing operations costs. The Storage Networking Industry Association (SNIA) is working on standards for ILM in these areas by extending the Storage Management Initiative Specification (SMI-S). Once completed, SMI-S standards for data, information, and security services will automate ILM practices.
SNIA’s Data Management Forum defines “tiered storage” and “ILM” as follows:
Tiered storage establishes a hierarchy of storage systems based on service requirements such as performance, business continuity, security, protection, retention, compliance, and cost. A company may implement a number of tiers, such as primary/secondary/backup/archive/compliance disk, tape-based backup/archive, and optical archive, etc.
ILM is described as a “standards-based, business-driven management practice,” or more simply as “information-based management.” ILM is not a specific service such as archiving or tiering; rather, it is about establishing business requirements for information and then via classification methods adhering to standard practices and service levels.
It is common to see the terms “tiered storage” and “ILM” used interchangeably. However, tiered storage is how repositories are arranged. Calling tiered storage “ILM” is analogous to saying backup is ILM or archiving is ILM. In a tiered storage context, ILM is how you manage the repositories based on requirements over time. ILM aligns business requirements and processes with service level requirements in an automated fashion, from data creation to deletion. Tiered storage should be viewed as one of many components encompassed by ILM.
Many applications have the ability to operate according to ILM-based practices, including tiered storage. You can start small and grow as your requirements change. Begin with a practice such as backup by implementing a backup array or virtual tape system connected to your tape automation systems, and let the backup media manager and virtual tape system automate the tiers. Or, implement an e-mail archive system that integrates multiple tiers for primary storage, archive, and/or compliance stores. Next, free up space on primary storage by deleting expired data or allow the database to archive it to secondary storage. Many databases, database archiving tools, and enterprise content management (ECM) platforms support automated, tiered storage.
Next, consider storage resource management (SRM) tools to profile your unstructured information. Find the active, inactive, reference, and expired data and then set up classes based on the value of the information to the business. Move inactive or expired information off primary storage.
Establishing tiered storage requires mechanisms to place data. There are three basic methods:
- Static, in which applications assign information to specific tiers;
- Staged, which refers to batched data movement (e.g. archive); and
- Dynamic, which requires active data movers (i.e., hierarchical storage management or ILM policy services).
The biggest challenge users face when they are beginning to implement ILM-based tiered storage is agreeing on the information and data classification requirements. This is the crux of establishing successful ILM practices. A successful ILM deployment begins with collaboration, where the objective is to have all relevant business groups participate in setting the requirements. This may require collaboration with records and information managers, security managers, and sometimes the legal and finance departments. We recommend adding intelligence to tiered storage by integrating it into a broader ILM-based practice. For more information, visit www.snia.org/dmf.
Michael Peterson is program director of the Storage Networking Industry Association (SNIA) Data Management Forum.