INTRODUCTION
Ten years ago, the computing power of a single Unix system barely met
the requirements of a typical application. This created a trend to use a
large number of smaller systems, each running their own discrete application.
Each system managed its own resources for its application. Today, however, the
typical server is many times larger, so we often encourage running multiple
applications on each system. Unix is a timeshare operating system and attempts to
distribute the resources it manages among the applications which it hosts.
However, the distribution of these resources does not always align with the
requirements of a given application. This often means that required service
and performance levels are not met due to conflicting requirements for the same
resources.
Resource Management is about managing system resources to optimize service
and performance levels, using methodologies that MVS Mainframe administrators
are familiar with. By using policies and resource management facilities, we
are able to provide control over how resources are allocated to applications.
The Resource Management BluePrint provides an introduction to resource
management principles and in depth details on implementating resource
management in the Solaris environment. It covers the various tools and techniques
that facilitate resource management, as well as how they should be implemented
to allow system managers to meet their service level and application performance
targets.
Specifically, the Resource Management BluePrint covers:
- Introduction to resource management methodology
- Analysis and application of resource management to workloads
- Measurement and monitoring in a resource managed environment
- Policies for resource management and Sun Microsystems products for Resource Management
- Base Solaris and processor sets
- Solaris Resource Manager
- E10000 system domains and dynamic reconfiguration
- Network Resource Management
- Network batch job management with LSF
- Symon
- Resource management on clusters
|
Close