Presentation is loading. Please wait.

Presentation is loading. Please wait.

Overview of Matchmaking and Scheduling (Mapping) Algorithm

Similar presentations


Presentation on theme: "Overview of Matchmaking and Scheduling (Mapping) Algorithm"— Presentation transcript:

0 MapReduce-Simulator: Matchmaking and Scheduling Algorithm
SYSC 5807: Resource Management on Distributed Systems MapReduce-Simulator: Matchmaking and Scheduling Algorithm Norman Lim Dept. of Systems and Computer Engineering Carleton University Ottawa, ON, Canada

1 Overview of Matchmaking and Scheduling (Mapping) Algorithm
1. When a job j arrives, it is placed in a queue depending on the scheduling policy selected. FIFO: uses first-in first-out queue EDF: uses earliest deadline first queue 2. Job/Task Mapping algorithm is invoked. Retrieves the first job in the queue and maps each of the job’s tasks onto the resources. Maps each task t onto a resource r that can execute t at its earliest possible time (at or after its earliest start time). For a particular job, the tasks in that job with higher execution time are mapped first.

2 Overview of Matchmaking and Scheduling (Mapping) Algorithm Cont.
EDF Policy Only 3. If a job j is not able to be scheduled to complete executing before its deadline, then the Job/Task Re-mapping algorithm is invoked. Remaps j and the set of jobs (called S) that have caused j to miss its deadline. S includes jobs that are scheduled to start at, or complete within the interval: [start time of job j, deadline of job j]. j and the jobs in S are re-ordered according to earliest deadline first. The Job/Task Mapping algorithm (see previous slide) is invoked to map j and the jobs in S.

3 Example: Jobs and Resources
Job 1 (j1) Arrival time = 0s EST = 2s Deadline = 30s t1 2 t2 5 t3 8 t4 10 t5 11 Job 2 (j2) Arrival time = 1s EST = 1s Deadline = 20s t1 4 t2 5 t3 10 Resource 1 (r1) Map Capacity = 1 Reduce Capacity = 1 Resource 2 (r2) Map Capacity = 1 Reduce Capacity = 1 Two jobs: j1 and j2. DAG shows the jobs’ tasks. - Blue circles: Map tasks - Red circles: Reduce tasks The task id and its required execution time are shown in the Two resources: r1 and r2. EST = earliest start time

4 Example: FIFO Job 1 (j1) Arrival time = 0s EST = 2s Deadline = 30s
FIFO Queue map j1, t2 j1, t1 j2, t2 j1, t4 j2, t3 reduce j2 will miss its deadline. Job 1 arrives Job 2 arrives Job 1 gets scheduled Job 2 gets scheduled. map j1, t3 j2, t1 j1, t5 reduce

5 Example: EDF (Case 1) Job 1 (j1) Arrival time = 0s EST = 2s
Deadline = 30s Job 2 (j2) Arrival time = 1s EST = 1s Deadline = 20s EDF Queue map j2, t1 j1, t3 j1, t5 reduce Job 1 arrives. Job 2 arrives, and goes to the start of the queue. Job 1 and 2 are scheduled. Note: In this case, Job 1 is not scheduled immediately because the scheduler was busy. map j2, t2 j1, t2 j1, t1 j2, t3 j1, t4 reduce

6 Example: EDF (Case 2) Job 2 (j2) Arrival time = 1s EST = 1s
Deadline = 20s Job 1 (j1) Arrival time = 0s EST = 2s Deadline = 30s Job 1 (j1) Arrival time = 0s EST = 2s Deadline = 30s Job 2 (j2) Arrival time = 1s EST = 1s Deadline = 20s EDF Queue map j2, t1 j1, t2 j1, t3 j1, t1 j2, t2 j1, t5 j1, t4 j2, t3 reduce j2 will miss its deadline. Job 1 arrives, and gets scheduled. Note: In this case, Job 1 is scheduled immediately because the scheduler was not busy. Job 2 arrives, and gets scheduled. Job 2 is not able to finish its execution before its deadline so the Job/Task Re-mapping algorithm is executed to remap the jobs j1 and j2. This time, Job 2 (earlier deadline) is scheduled first, followed by Job 1. map j2, t2 j1, t3 j1, t2 j2, t1 j1, t1 j2, t3 j1, t5 j1, t4 reduce


Download ppt "Overview of Matchmaking and Scheduling (Mapping) Algorithm"

Similar presentations


Ads by Google