InformationWeek.com

Join Kevin Mitnick - the "most wanted computer criminal in the world" - as he shares his secrets on the security threats facing corporate America.

Welcome Guest. | Log In| Register | Membership Benefits

December 7, 1998

Print this story

Methods For Management

IT departments must put processes in place to ensure high levels of application availability

By Roger Fournier

s competition increases in the global economy, the effective use of business-enabling technology is no longer just a competitive advantage. It's a fundamental necessity to ensure business growth.

Companies are deploying growing numbers of critical applications that are tightly linked with core business processes. These applications require round-the-clock reliability, and IT organizations must create a robust distributed infrastructure that provides high levels of serviceability, performance, availability, reliability, and supportability.

Technology is just one ingredient required to ensure high levels of application availability. IT's challenge is to provide the right mix of technology, people, processes, and organizational structure. IT managers must successfully plan, implement, and support a set of core distributed-systems-management processes necessary to meet or exceed users' expectations. This includes best practices for enterprise system management.

Business Alignment
The identification of a company's core business needs is the starting point for laying down a solid foundation upon which IT can implement comprehensive systems-management disciplines, on top of its underlying technology infrastructure. Understanding customers' core requirements for quality service is critical for creating that foundation.

IT infrastructure services and business application architecture must be closely aligned with an enterprise's strategic goals and objectives and the set of critical business processes that are aimed at supporting these. Here's a brief description of some of the most common systems-management processes needed to successfully support complex, heterogeneous distributed computing environments.

Antivirus-protection management entails all activities required to prevent software destruction and the loss of crucial data due to virus attacks. Effective virus-protection mechanisms are usually managed through a centralized administration and control process. Real-time antivirus mechanisms must protect a variety of IT assets, including client software, servers, networks, intranets, and electronic messaging systems. Antivirus-management practices must be closely integrated with other enterprise-security-management practices.

Application management encompasses all activities required to manage key business applications and commercial packages. This includes the identification, monitoring, tracking, and deployment of software components across an enterprise's distributed environment. The monitoring process is performed by agents, which track specific application-related thresholds and continually gauge the overall performance of the business systems. A major benefit of effective application management is increased application availability, which directly translates to enhanced user productivity. Application management is particularly popular among companies that want better control over large and complex enterprise resource planning applications.

Asset management encompasses all activities related to inventorying technology assets such as networks, software, and hardware. It also entails controlling and regulating the use of configuration files. Finally, asset management includes monitoring and auditing software usage across the enterprise. Accurate and up-to-date asset-management practices are useful to perform help-desk functions and support the distribution of software. Asset management is tightly coupled with configuration management.

Backup and recovery management includes all activities required to back up key business databases and applications. It also involves recovering databases and applications as fast as possible. Best practices include the ability to back up databases online, without having to bring the database offline.

Capacity planning encompasses the set of activities that are used to proactively collect, consolidate, and analyze client-server and Web application-performance data. Capacity planning's prime objective is to better predict the planned utilization of distributed computing resources such as servers, operating systems, networks, and middleware. The most common resources assessed include memory capacity, CPU power, input/output transfer rates, network bandwidth, and disk storage. Capacity-planning management has strong dependencies with the management of performance monitoring and tuning, as well as service-level management.

Change management entails all activities related to the planning, coordination, and implementation of hardware, network, and software changes. It entails assessing the impact of proposed changes and notifying all affected users of upcoming operational changes. Finally, it also requires the development of a contingency plan to restore the production environment to its previous state should changes not work properly.

Change management is one of the most fundamental systems-management disciplines that must be enforced to ensure robustness--serviceability, performance, availability, reliability, and supportability--in a distributed computing environment. Change-management best practices encompass the control of all changes made in the distributed production environment, including operating systems, software upgrades and modifications, applications, application changes to online transactions and batch jobs, application transactions, batch jobs additions and deletions, changes to hardware configurations, and network modifications such as performance-related enhancements. Change management is closely tied to configuration management.

Change-management procedures should clearly spell out the roles and responsibilities of each group involved in the change-management process. Furthermore, the change-management procedures should cover both scheduled and emergency changes.

Configuration management covers all the activities that relate to discovering, classifying, regulating, and monitoring the content and version of crucial hardware, network, and software components, along with their interdependencies. Good configuration-management practices help the IT department conduct thorough impact assessments when the change-management group must coordinate production changes of critical infrastructure components, such as production servers. Configuration management is also closely associated with asset and change management.