HP Integrity Essentials Global Workload Manager: Workload Management for HP Integrity Virtual Machines

3
As a result, you can easily make more efficient and effective use of your Integrity VM resources.
While maintaining your targeted business priorities and service levels, you can consolidate more
workloads onto fewer physical systems, reducing the physical hardware required and thereby
significantly reducing costs.
Intelligent, Automatic, and Diverse Resource Management Capabilities
Mapped to Your Business Priorities
Both HP Integrity VM and gWLM provide mechanisms for dynamic resource allocation. The HP
Integrity VM includes a fixed, built-in policy (entitlement) that determines a minimum percentage of
physical CPU (cores) to be allocated for a virtual machine, based on the number of virtual CPUs
(vCPUs) configured. (Note that HP Integrity VM documentation generally uses the term “physical CPU”
instead of “core” to help distinguish between physical and virtual components.) If a virtual machine is
busy and sufficient processing power is available on the VM Host system, the virtual machine can
receive more than its entitlement. When there is contention for processing power among the virtual
machines, each virtual machine is limited to its entitlement. As needed, you can manually change
these entitlements dynamically (while the Integrity VM software and the applications or processes it
serves are running) to address needs for reallocating resources. You cannot set a single entitlement to
be applied to all the virtual machines. The Integrity VM software is not capable of activating or
deactivating Instant Capacity resources as needed.
In contrast, gWLM includes intelligent, automatic resource management capabilities and provides you
more extensive and varied resource allocation options so that you can control the behavior of your
virtual machines to align more closely to your business needs. For example, if you want to consolidate
multiple virtual machines that run workloads having different business priorities, gWLM can
automatically change CPU resource allocations to individual virtual machines based on these
priorities. gWLM can also maintain targeted utilization goals and service levels for each virtual
machine.
gWLM manages resources for workloads based on policies that you define and apply to those
workloads. You can establish multiple policies (individualized for particular virtual machine
workloads) or you can establish a single policy for all the virtual machine workloads to minimize the
number of policies. To establish a gWLM policy, you can choose from a wide variety of rules for
allocating and sharing resources. For example, besides being able to assign the allocation owned by
a virtual machine workload, where the owned allocation is similar to an Integrity VM entitlement, you
can also assign a minimum allocation that can be greater than the Integrity VM configured resource
minimum, and you can assign a maximum allocation that can be lower than the virtual machine’s
configured size. Thus, you can set a limit for the amount of resources a virtual machine can use
without having to change its configuration. In addition, gWLM allows you to assign a priority level
and weight to each virtual machine workload, helping ensure that critical applications get the
resources they need. As business priorities and conditions change, you can easily modify your
policies to reflect the changing needs. gWLM then automatically reassigns resources to your
workloads to improve resource utilization and maintain continuous service levels. For more
information on gWLM policies relevant for managing Integrity VM, see “gWLM Policies You Can
Establish for Virtual Machine Workloads” on page 6.
Intelligent and Automatic Activation/Deactivation and Distribution of
iCAP or TiCAP Resources Based on Business Priorities
For Integrity VMs deployed on cellular, partitionable servers, gWLM can automatically activate or
deactivate Instant Capacity (iCAP) and Temporary Instant Capacity (TiCAP) resources available to the
VM Host; you pay for these resources only as needed. gWLM can instantly activate these resources to
increase capacity to accommodate increased demands of virtual machine workloads or to replace
failed physical CPU resources on the VM Host. gWLM distributes the activated resources according to
your business priorities. They can be activated without changes to the virtual machine configurations;
the virtual machines immediately take advantage of the additional processing power. Note that