ILM advice: Collaboration is critical

Successful implementation of information lifecycle management starts with collaboration among various IT and non-IT groups within an organization.

By Michael Peterson

If you are a CIO or information management professional, your job may be at risk. This is one of the messages in a recent paper, Collaboration: the News Standard of Excellence, jointly written by ARMA International and the Storage Networking Industry Association (SNIA).

The message of “collaboration” should be taken holistically. Perhaps the hardest step in the process of implementing practices such as information life cycle management (ILM) is to bring the information owning and administrating communities of an organization together to collaborate and align business objectives with the requirements for information. Although challenging, this is the most important place to start, because this is where the entire transformational process of moving to an “information-centric enterprise” begins.

The underlying collaboration message is the following: If you want to successfully solve the complexity and cost crisis in data-center operations, you have to change current practices and begin working together as an organization. If you want to successfully deal with the demands of regulatory compliance and legal and security risk, you have to focus on defining the business requirements for your information and not just focus on the network, security, or storage infrastructure.

SNIA’s contention in the collaboration paper is that you must make the shift to managing and operating the data center based on the value of, and requirements for, information to the organization. Step one in information-based management is to collaborate and define the requirements for information so that it can be managed properly over its lifecycle. This is the key to ILM.

For those who have grown tired of the ILM story, it’s important to understand that ILM is just coming of age. We are only three years into the market development cycle and already second-generation tools are in the market. Thousands of organizations worldwide are in various stages of ILM projects and their numbers are growing rapidly. ILM is no longer just something that vendors talk about as a way to promote their products; it is a proven practice.

The evolution of ILM is not much different from that of other storage technologies. For example, SANs were launched in 1996, work on Fibre Channel standards began in 1997, and market acceptance took off in 1999-three years after the initial launch. We’re now 10 years into the market cycle and SANs are common. Similarly, iSCSI began around 2000, market acceptance followed a few years later, and we are now in year eight of a growing market. The process always follows a similar timeline, and ILM is no exception.

What is the connection between collaboration and ILM, and what does a complete ILM-based practice look like? Consider a market map, which is an illustration that aids in visualizing how all the pieces of a market fit together. For example, markets begin with user-centric drivers of demand, and solutions are then identified that address those needs. A market map for ILM-based practices is illustrated in the figure. Here is how the pieces fit together:

Click here to enlarge image

The leading drivers for using information as the basis for management (and to deal with overwhelming cost and complexity) are elements of what we call information convergence, which is the trend in which operations, practices, applications, and roles converge around information and its value to the organization, creating the information-centric enterprise. The top information drivers are regulatory compliance, legal discovery risk, and security risk. Storage problems such as “out-of-control capacity growth” are not drivers; they are results of the drivers-the need to store more, retain it for longer periods, secure it, protect it, and retain more copies in multiple locations, etc.

Organizational response transforms the organization into an information-centric enterprise-an organization in which the value of, and requirements for, information are used as the central basis for management, administration, and security operations. This creates a collaborative relationship among information users, information owners, and information systems operators and administrators.

The best organizational response to an information convergence problem is collaboration among the communities that have a vested interest in the information: the business group, legal, finance, records and information managers, and IT and security professionals. The purpose of collaboration is to define requirements for information and information resources.

Collaboration is the first step in an information-based management approach, and the second step is information classification.

Information classification is the process of assigning value and scope to a business process. It is the process of aligning the value and requirements for information with the supporting infrastructure resources (e.g., compute, network, storage, and supporting data, security, and storage services).

The next step is to implement ILM-based practices by developing a comprehensive management practice around the requirements for information based on its classification. This is the foundation for executing ILM-based practices-to translate specific business and application policies and rules (service level objectives and agreements) into actionable practices. Tools are available to support these practices, with capabilities such as the following:

  • Classifying data based on file metadata or content;
  • Assigning service-level requirements to the data based on its classification;
  • Classifying storage resources based on performance, availability, or other service level attributes; and
  • Aligning the data with the most-effective storage resources.

Applications and ILM management tools implement and automate the policies for their respective domains using available services. These data, security, and storage services, as illustrated in the figure, right, provide the tools and technologies by which IT delivers the infrastructure to meet service-level requirements that can flexibly change over the data’s lifecycle.

The next stages of market maturation for ILM-based practices have two dimensions:

  • Introduction of better automation and management tools, with centralized management platforms and practices integrated with enterprise applications; and
  • Release of the initial specifications for two standards being developed by SNIA, the ILM-related portions of SMI-S and XAM, and their integration into products.

The first of these standards is SNIA’s Storage Management Initiative-Specification (SMI-S)for ILM services. Just as SMI-S has provided standard management interfaces for managed devices and network elements, SMI-S for ILM services will allow heterogeneous services to be instrumented and driven by central ILM management tools and applications. SMI-S-based instrumentation of services is essential for automation of ILM-based practices. The first such specifications will appear in SMI-S 1.2.0, with more management interfaces to follow in subsequent releases.

The second standard is the eXtensible Access Method. XAM, which is currently being developed by SNIA, provides applications with a standard interface to storage, beginning with object storage systems such as content-aware storage (CAS), with the capability to write metadata relevant to ILM practices. As XAM becomes incorporated in a broad spectrum of applications, it will become strategic to ILM-based automation.

Although your ability to instrument and automate your infrastructure and services will be limited until these standards are completed and adopted, there are many practices you can implement today. For example, you can build and automate solution stacks for ILM-based practices around applications such as e-mail or database archiving. You can implement tiered storage, data protection, compliance, and archiving solutions, and automate them using either ILM management tools or utilities integrated into virtualization platforms.

Click here to enlarge image

The most-effective way to get started with ILM is to have the CFO or CIO engage with a professional services organization for an ILM assessment of the effectiveness of data-center processes. This engagement also allows the professional services organization to engage with the data center, lines of business, and key enterprise stakeholders such as legal, security, and records and information managers to help your organization take the first steps of collaboration and information classification.

For more information, visit the SNIA Data Management Forum’s Website at www.snia-dmf.org.

Michael Peterson is the program director of the SNIA Data Management Forum and president of Strategic Research Corp. (www.sresearch.com).

This article was originally published on December 01, 2006