skill-tree:k:4:b
Differences
This shows you the differences between two versions of the page.
Both sides previous revisionPrevious revisionNext revision | Previous revision | ||
skill-tree:k:4:b [2020/06/25 17:10] – [Background] kai_h | skill-tree:k:4:b [2020/07/19 19:44] (current) – lucy | ||
---|---|---|---|
Line 1: | Line 1: | ||
# K4-B Job Scheduling | # K4-B Job Scheduling | ||
# Background | # Background | ||
- | Parallel computers are operated differently than a normal PC, all users must share the system. Therefore, various operative procedures are in place. Users must understand these concepts and procedures to be able to use the available resources of a system to run a parallel application. | + | Parallel computers are operated differently than a normal PC, all users must share the system. |
+ | Therefore, various operative procedures are in place. Users must understand these concepts and procedures to be able to use the available resources of a system to run a parallel application. | ||
A workload manager/job scheduler controls how available hardware resources are distributed among the user requests (jobs). | A workload manager/job scheduler controls how available hardware resources are distributed among the user requests (jobs). | ||
- | Users of compute | + | Users of computing |
- | HPC resources can be distinguished as | + | HPC resources can be distinguished as |
- | * shared | + | * Shared |
- | * not-shared resources (e.g. cluster nodes dedicated to a particular parallel program of an individual user). | + | * Not-shared resources (e.g. cluster nodes dedicated to a particular parallel program of an individual user). |
- | | + | |
The configuration of the cluster system matters as well: a cluster node can also be a resource that is shared between several users. | The configuration of the cluster system matters as well: a cluster node can also be a resource that is shared between several users. | ||
- | A major aspect of job scheduling is to manage these resources in a way that users are treated fairly. | + | |
+ | A major aspect of job scheduling is to manage these resources in a way that users are treated fairly. | ||
Accounting for users or user groups can additionally support this. | Accounting for users or user groups can additionally support this. | ||
# Aim | # Aim | ||
- | * To enable practitioners to comprehend and describe the basic architecture and concepts of resource allocation for an HPC system | + | * To enable practitioners to comprehend and describe the basic architecture and concepts of resource allocation for an HPC system. |
- | * To provide | + | * To provide |
- | * To provide | + | * To provide |
# Outcomes | # Outcomes | ||
- | * comprehend | + | * Comprehend |
- | * explain | + | * Explain |
- | * run interactive jobs and batch jobs | + | * Run interactive jobs and batch jobs. |
- | * comprehend | + | * Comprehend |
- | * change | + | * Change |
- | * analyze | + | * Analyze |
- | * comprehend | + | * Comprehend |
- | * comprehend | + | * Comprehend |
- | * Resource Utilization | + | * Resource Utilization. |
- | * Throughput | + | * Throughput. |
- | * Waiting Time | + | * Waiting Time. |
- | * Execution Time | + | * Execution Time. |
- | * Turnaround Time | + | * Turnaround Time. |
- | * comprehend | + | * Comprehend |
- | * comprehend | + | * Comprehend |
- | * maximization | + | * Maximization |
- | * maximization | + | * Maximization |
- | * minimization | + | * Minimization |
- | * minimization | + | * Minimization |
+ | * Comprehend that there is a variety of scheduling algorithms from rather simple to more complex like: | ||
+ | * First-Come-First-Served (FCFS). | ||
+ | * Shortest-Job-First (SJF). | ||
+ | * Priority-based. | ||
+ | * Fair-Share. | ||
+ | * Backfilling. | ||
# Subskills | # Subskills | ||
* [[skill-tree: | * [[skill-tree: | ||
* [[skill-tree: | * [[skill-tree: | ||
+ | * [[skill-tree: | ||
+ | * [[skill-tree: |
skill-tree/k/4/b.txt · Last modified: 2020/07/19 19:44 by lucy