HP-UX Workload Manager

Rambo

HP-UX
Workload Manager (HP-UX WLM) is a resource management tool that
provides automatic CPU resource allocation and application performance
management based on your service-level objectives (SLOs) and business
priorities. WLM is most effective managing applications that are
CPU-bound. It automatically adapts system or partition CPU resources
(cores) to the demands, SLOs, and priorities of the running
applications. (A core is the actual data processing engine within a
processor, where a single processor can have multiple cores, and a core
can support multiple execution threads through Hyper-Threading,
introduced with HP-UX 11i v3, as discussed below.) It adjusts the CPU
allocation of a group of processes known as a workload, basing
adjustment on the current needs and performance of the applications in
that workload. A workload is based on the collection of processes in:

nPartitions that use Instant Capacity

HP-UX virtual partitions

HP-UX Virtual Machines (Integrity VM) hosts

Resource partitions, which can be:

Whole-core: HP-UX processor sets (pSets)

Sub-core: Fair Share Scheduler (FSS) groups (WLM creates FSS groups using HP Process Resource Manager (PRM))
Note
As of WLM A.03.01, PRM is no
longer included with the WLM bundle. If PRM C.03.00 or later is already
on the machine on which you must install or upgrade WLM, you can
continue to manage FSS and pSet-based workload groups just as if PRM
had been installed with WLM. If you are installing WLM for the first
time on a machine, you can use a strictly host-based configuration (no
FSS or pSet workload groups). However, to manage FSS and pSet-based
workload groups, you must install PRM (C.03.00 or later) separately.
With your purchase of WLM, you are entitled to a PRM license. For
information on receiving the PRM license, contact your regional
licensing service organization.







»
HP-UX Workload Manager
»
Overview and Features
»
Quickspec
»
Information library
»
Platforms
»
Frequently asked questions
»
Case studies
»
Example configurations
»
Patches / support
»
Documentation corrections




»
Contact HP
»
Download WLM / WLM toolkits




»
Training
»
User Forum








WLM is optimized for
moving cores among hosts such as virtual partitions and nPartitions.
Using these hosts as workloads, WLM manages workload allocations while
maintaining the isolation of their HP-UX instances. WLM automatically
moves or "virtually transfers" cores among partitions based on SLOs and
priorities that you define for the partitions. WLM can manage nested
workloads, with workloads based on FSS groups and pSets inside virtual
partitions inside nPartitions.
You can also use WLM to manage workloads to divide resources within
a single HP-UX instance. In this case, you manage SLOs for workloads
based on PRM-based pSets or FSS groups. These workloads are usually
referred to as "workload groups." WLM can manage real memory and disk
bandwidth within an HP-UX instance, although not in response to SLO
performance. Disk bandwidth can be statically allocated.
WLM supports the logical CPU (Hyper-Threading) feature for
pSet-based groups. Hyper-Threading is available on certain processors
starting with HP-UX 11i v3 (B.11.31). A logical CPU is an execution
thread contained within a core. Each core with Hyper-Threading enabled
can contain multiple logical CPUs. WLM automatically sets the
Hyper-Threading state for the default pSet to optimize performance.
(The default pSet is where FSS groups are created.) When new pSets are
created, they inherit the Hyper-Threading state that the system had
before WLM was activated (because WLM may change the Hyper-Threading
setting of the default pSet to optimize performance). Cores can be
moved from one partition to another and will take on the
Hyper-Threading state of their destination pSet. You can override the
default state for cores assigned to a specific pSet-based group; you
can also modify the Hyper-Threading state of the system. (Modifications
to the Hyper-Threading state should not be made while WLM is running.)
Key uses of HP-UX WLM include:

Using
excess server capacity by consolidating multiple applications on fewer
servers while ensuring that mission-critical applications still get the
resources they need in times of peak demand

Automatically reallocating system resources in response to changing
priorities, conditions that change over time (night/day, month-end
processing, and so on), package movement in a cluster, resource demand,
and application performance

Automating the deployment of reserve capacity so that customers pay only for what they need when they need it

Enabling higher utilization in clusters by enabling you to define,
monitor, and enforce SLOs on a server or partition that receives a
failed over workload You can use WLM within a whole server that can be clustered in an
HP Serviceguard high availability cluster, Extended Campus Cluster,
Metrocluster, or a Continentalcluster configuration. You can also use
WLM on an Integrity VM host and within any individual Integrity VM
(guest).  HP also provides Workload Manager Toolkits (WLMTK),
which simplify the implementation of HP-UX WLM and allow WLM to easily
integrate with third-party applications. You can download WLMTK for
free, as well as learn more about it from its user's guide in the
information library. The implementation of HP-UX WLM is
application-transparent: You do not need to modify your applications to
manage them with WLM. HP-UX WLM provides features and solutions such as:

Service-level objectives
With
HP-UX WLM, you define objectives with a priority, which you then assign
to a WLM workload. You can set objectives to meet utilization goals or
metric goals. (Metric goals are based on performance data and require
understanding of that data. HP recommends using utilization goals,
which can be implemented immediately without prior knowledge of
workload performance.) Also, you can set objectives to provide a fixed
allocation or a shares-per-metric allocation, which for example, can
give a workload 2% of the CPU resources for each process in the
workload. You can assign one or more SLOs to a workload. The
ability to assign multiple SLOs is helpful for workloads that require
more than one SLO to accommodate a "must meet" goal and optional, lower
priority stretch goals. You can monitor your SLOs through either a command-line interface or a graphical interface.

Automatic CPU allocation
HP-UX WLM
automatically allocates CPU resources to maintain application
performance during changing system conditions and fluctuations in
workload demand.

Secure Resource Partitions
HP-UX WLM supports
Secure Resource Partitions
,
which combine HP-UX 11i V2 security capabilities with the resource
management capabilities of HP-UX WLM. This combination allows you to
consolidate multiple applications within a single operating system
image, providing them dedicated resources plus operating system level
security from other applications.

Automatic management of SLOs globally, across multiple systems
HP-UX
Virtual Partitions (vPars) are software-based virtual systems, each
running its own instance of the HP-UX operating system. WLM can move
processors among these virtual systems to better achieve the SLOs you
define using WLM within each virtual system. nPartitions are
hardware-based partitions, each running its own instance of the HP-UX
operating system. With Instant Capacity (iCAP) software (formerly known
as iCOD software) installed on each nPartition, WLM can simulate
movement of cores by deactivating a core on one nPartition then
activating a core on another nPartition, moving the resources to where
they are most needed. You can use virtual partitions within
nPartitions. Virtual machines, like virtual partitions, are
created by software. Virtual machines emulate generic servers, and
therefore can offer sub-core and shared I/O capabilities. Each virtual
machine runs its own operating system. HP Integrity Virtual Machines
can be used within hard partitions.

System consolidation
With HP-UX WLM,
because each application is given only what it needs, when it needs it,
the excess capacity is shared more efficiently.

Passive-mode operation
WLM provides a
passive mode that allows you to see how WLM will approximately respond
to a given configuration — without putting WLM in charge of your
system's resources. Using this mode, you can analyze your
configuration's behavior — with minimal effect on the system. Besides
being useful in understanding and experimenting with WLM, passive mode
can be helpful in capacity-planning activities.

Response-time management
You can improve the response time for critical users and applications through effective use of workloads and SLOs.

Performance expectation
You can set and manage user expectations for performance with workload and goal definitions.

Billing information
WLM allows you to
generate billing information on the system level and on the partition
level (virtual partition or nPartition).
HP-UX WLM integrates with numerous products to allow you to better manage your systems. For example, you can use WLM with:

An HP Serviceguard clustering solution
When
HP Serviceguard moves an application from one server to another, HP-UX
WLM immediately adjusts the resources to guarantee the right resources
for the moved application, based on defined SLOs. For more information,
read the
"Serviceguard-WLM integration"
white paper.

Temporary Instant Capacity (TiCAP) and Pay per use (PPU)
You
can use WLM to manage TiCAP and PPU resources to ensure that your
workloads use only the number of cores needed to meet the workloads'
SLOs and that you pay only for what resources you actually use.
Temporary Instant Capacity activates capacity in a temporary
"calling-card fashion" such as in 30-day increments where a day equals
24 hours for one core. With this option, you can activate and
deactivate cores, optimizing the amount of time these resources are
used to meet the needs of your workloads. You purchase a codeword to
obtain rights to use certain Instant Capacity cores for a preset amount
of days. This codeword is applied to a system so that you can turn on
and off any number of these cores as long as your prepaid amount of
temporary capacity days has not expired.  WLM supports TiCAP version 6
or later.
Using WLM on a system with the PPU software, CPU
capacity is increased or decreased automatically to support peak
anticipated demand, basing payment for the HP server on actual metered
or monitored usage of that capacity. With PPU version 4, capacity can
be increased or decreased by whole cores as needed, with billing
determined by the number of active cores. Beginning with PPU version 5,
all cores on a PPU system are active and billing is based on your
percentage of usage of those cores. Starting with PPU version 7, which
includes version 5 capabilities, billing can also be based on the
number of active cores on the system, with WLM activating only those
cores that are needed. WLM integrates with PPU versions 4 and 7, or
later.

Oracle® instances
You
can control Oracle database instances, adjusting their CPU allocations,
based on a number of factors. These factors include desired transaction
response time, number of users connected, and whether a particular job
is active, among others.

SAP® instances and processes
WLM and
its SAP Toolkit (SAPTK), in conjunction with the HP Serviceguard
Extension for SAP (SGeSAP) product, take advantage of the WLM process
map feature to enable you to identify different SAP processes or
instances and place them into separate workloads. WLM can prioritize
and assign specific SAP processes to workloads.
For information on other products with which WLM integrates, see the Integration appendix in the HP-UX Workload Manager User`s Guide. HTML and PDF versions of this document are available at:
http://docs.hp.com/hpux/netsys/
. HP offers a 90-day trial version of HP-UX WLM. For more information, see the
HP Software Depot
.


»
Download the latest Adobe Acrobat Reader
               
               
               

本文来自ChinaUnix博客，如果查看原文请点：http://blog.chinaunix.net/u1/39440/showart_1136524.html