The gWLM Management Model

gWLM enables utility computing across a data center by providing resource-sharing policies that you centrally create and monitor. gWLM moves resources among the workloads in a shared resource domain (SRD) as needed—based on the policies you specify.

gWLM allows you to manage resource allocations for several types of system divisions, as discussed below. These divisions are referred to as compartments in gWLM.

HP-UX Hardware Partitions (npar)
A hardware partition, also known as an nPartition or npar, is a physical partition of a server, where each npar runs its own instance of the HP-UX operating system (which can include use of HP Integrity virtual machines) or is divided into virtual partitions.
Using the HP Instant Capacity product, gWLM simulates the movement of CPU resources among npars by turning off an active core in one npar then turning on a deactivated core in another npar in the same complex. Thus, the first npar has one less active core, while the second npar has one additional active core. (gWLM maintains the number of active cores, honoring the Instant Capacity usage rights. As a result, no additional costs are incurred.) Combining gWLM A.04.00.07 or later and an appropriate version of the iCAP software, gWLM's ability to manage npars using iCAP is extended across multiple complexes that are members of the same Global iCAP group.
HP-UX Virtual Partitions (vpar)
A virtual partition is a software partition of a server or of a single nPartition, where each virtual partition runs its own instance of the HP-UX operating system. A virtual partition cannot span an nPartition boundary.
HP Integrity Virtual Machines (hpvm)
Virtual machines are a robust soft-partitioning and virtualization technology that provides operating system isolation, with sub-core allocation granularity and shared I/O. These virtual machines can run a variety of operating systems. gWLM can manage a virtual machine regardless of the operating system running inside it.
Processor sets (psets)
A processor set is a collection of cores (formerly known as CPUs) grouped together for the exclusive access by processes assigned to that processor set. Processor sets form partitions within a single operating system image.
HP-UX Fair Share Scheduler groups (fss groups)
A group of processes that has its CPU resource allocation managed by the Fair Share Scheduler that is available with HP-UX. A benefit of fss groups is their granularity: You can allocate fractions of CPU resources, rather than only whole cores, to the group of processes. These groups form partitions within a single operating system image.

For more information on these partitions, visit:

HP Virtual Server Environment web site:
http://www.hp.com/go/vse
The “Technical Documentation web site for HP Virtual Server Environment (VSE)” web site:
http://docs.hp.com/en/vse.html
The “Global Workload Manager” topic and the glossary in the online help for gWLM, available in gWLM’s graphical interface in HP SIM.

gWLM manages resources based on the following model:

You define an SRD by:
1. Deciding which of your systems you want to manage and what type of compartments you want to use. (gWLM manages existing npars, vpars, and virtual machines. It can manage your existing psets as well as create new ones. It creates fss groups for you.)
2. Associating each workload with a compartment. (For npars, vpars, and virtual machines, the compartment itself defines the workload. For psets and fss groups, you define the workload based on applications, users, or process IDs.)
3. Associating a policy with the workload indicating how gWLM should allocate resources to the workload's compartment. (gWLM comes with several policies and also lets you define your own. You can use a single policy for multiple workloads, minimizing the number of policies, if desired.)
Once the SRD is deployed:
1. gWLM monitors the CPU resource consumption of all the workloads in the SRD during the current allocation interval.
2. At the end of the interval, gWLM adjusts the CPU resource allocations for the compartments in accordance with the policies. It also makes the allocation data available for real-time and historical reports.
3. gWLM repeats the previous two substeps.

For information on what types of workloads to combine for optimal resource utilization, refer to the online help topic “Getting the Most Out of gWLM.”

The gWLM Management Model

Technical documentation

» Table of Contents

» Index