CPUSLICE parameter: Difference between revisions
(Automatically generated page update) |
mNo edit summary |
||
| Line 1: | Line 1: | ||
{{Template:CPUSLICE parameter subtitle}} | {{Template:CPUSLICE parameter subtitle}} | ||
==Summary== | ==Summary== | ||
<dl> | <dl> | ||
| Line 13: | Line 14: | ||
<dd><var class="product">Model 204 V6.1</var> or earlier | <dd><var class="product">Model 204 V6.1</var> or earlier | ||
</dl> | </dl> | ||
==Description== | ==Description== | ||
<p>The CPU time slice allotment in milliseconds for CPU-bound users</p> | <p> | ||
<p>If a CPU-bound user does not yield the CPU voluntarily after CPUSLICE milliseconds, the user is timesliced, that is, forced to yield the CPU to the user with the next highest priority. </p> | The CPU time slice allotment in milliseconds for CPU-bound users</p> | ||
<p>CPU-bound users that voluntarily yield the CPU are again considered IO-bound. For more information about IO-bound and CPU-bound, see [[IOSLICE parameter|IOSLICE: CPU time slice allotment]].</p> | <p> | ||
<p>Increasing this parameter results in improved service for CPU bound requests and reduced service for other requests. A smaller value for this parameter increases the rate at which a CPU-bound request falls in priority, resulting in poorer service for the request. </p> | If a CPU-bound user does not yield the CPU voluntarily after <var>CPUSLICE</var> milliseconds, the user is timesliced, that is, forced to yield the CPU to the user with the next highest priority. </p> | ||
<p>Although a small value for this parameter improves service for non-CPU-bound requests, it can also reduce overall system throughput and performance by increasing scheduler overhead. | <p> | ||
<p>For more information about <var class="product">Model 204</var> dynamic dispatching, see | CPU-bound users that voluntarily yield the CPU are again considered IO-bound. For more information about IO-bound and CPU-bound, see [[IOSLICE parameter|IOSLICE: CPU time slice allotment]].</p> | ||
<p> | |||
Increasing this parameter results in improved service for CPU bound requests and reduced service for other requests. A smaller value for this parameter increases the rate at which a CPU-bound request falls in priority, resulting in poorer service for the request. </p> | |||
<p> | |||
Although a small value for this parameter improves service for non-CPU-bound requests, it can also reduce overall system throughput and performance by increasing scheduler overhead. </p> | |||
<p> | |||
For more information about <var class="product">Model 204</var> dynamic dispatching, see [[Controlling system operations (CCAIN)#Dynamic dispatching|Dynamic dispatching]]. </p> | |||
[[Category:Scheduler parameters]] | [[Category:Scheduler parameters]] | ||
[[Category:System parameters]] | [[Category:System parameters]] | ||
[[Category:Parameters]] | [[Category:Parameters]] | ||
Latest revision as of 23:36, 17 November 2014
CPU slice - CPU
Summary
- Default value
- 10
- Parameter type
- System
- Where set
- On User 0's parameter line or reset by system manager
- Related products
- All
- Introduced
- Model 204 V6.1 or earlier
Description
The CPU time slice allotment in milliseconds for CPU-bound users
If a CPU-bound user does not yield the CPU voluntarily after CPUSLICE milliseconds, the user is timesliced, that is, forced to yield the CPU to the user with the next highest priority.
CPU-bound users that voluntarily yield the CPU are again considered IO-bound. For more information about IO-bound and CPU-bound, see IOSLICE: CPU time slice allotment.
Increasing this parameter results in improved service for CPU bound requests and reduced service for other requests. A smaller value for this parameter increases the rate at which a CPU-bound request falls in priority, resulting in poorer service for the request.
Although a small value for this parameter improves service for non-CPU-bound requests, it can also reduce overall system throughput and performance by increasing scheduler overhead.
For more information about Model 204 dynamic dispatching, see Dynamic dispatching.