Event Based Simulation of The Backfilling Algorithm

1
Event Based Simulationof The Backfilling
Algorithm

• OOP tirgul No. 4
• 2006

2
Outline

• The CPU allocation task revisited
• The Backfilling algorithm
• Simulation implementation issues

3
The CPU Allocation Task

• Task components
• Parallel computing machine with N CPUs
• A stream of request for running jobs
• Input for each job
• Id, arrival time, num of CPUSestimated run time
• The task
• Continuously decide which jobs to start and when
  to start

4
First Come First Served Revisited

• Simple algorithm
• Non optimal resource usage

processors
time
now
5
Exploiting Free Resources

• Start small tasks right now
• Do not delay tasks ahead in the queue

processors
time
now
6
Aggressive Backfilling (EASY)

• Start small tasks right now
• Do not delay the first task in the queue
• Runs on real parallel machines

processors
time
now
7
EASY Scheduler - Overview

• Main data structures
• waitQueue a queue of waiting jobs
• runList a list of running jobs
• Calculate status parameters
• shadow time
• free CPUs
• Find a backfill job

8
Data Structures

• WaitQueue
• Sorted according to arrival time
• For each job maintain
• Number of requested CPUs
• Estimated run time
• RunList sorted by termination time

processors
time
now
1st
2nd
3rd
4th
9
The Status Parameters

• shadow timeThe expected time in which the first
  job in the queue can run assuming no delays
• free CPUsThe minimum of
• number of expected free CPUs after 1st job in
  queue starts
• the CPUs that are currently free

Shadow time
free CPUs (in the initial stage)
processors
time
now
1st
2nd
3rd
4th
10
Calculating Status Parameters

• set expCapacity capacity // current free CPUs
• while expCapacity lt firstJob.numCPUs
• currJob next job in runList
• expCapacity expCapacity currJob.numCPUs
• shadowTime currJob.ExpectedFinishTime
• freeCPUs min ( expCapacity-firstJob.numCPUs ,
• capacity )

Shadow time
free CPUs (in the initial stage)
processors
time
now
1st
2nd
3rd
4th
11
Decision to Run a New Job Including Backfilling

• Loop on jobs in waitQueue in arrival order
• if ( job.numCPUs gt capacity)
• continue
• else
• if ( job is 1st in waitQueue
• estTermTimeIfStartsNow lt shadow
• job.numCPUs lt freeCPUs )
• start job and update freeCPUs

Shadow time
free CPUs (in the initial stage)
processors
time
now
1st
2nd
3rd
4th
12
Simulation of a Scheduler

• Goal
• analyze different schedulers
• Use records of running processes
• Input for each job
• Traces (log file) of real processes running on
  real machines
• Id, arrival time, num of CPUSestimated run time,
  actual runtime
• Output
• Various statistics on waiting time for each
  algorithm (average wait time and slow down in our
  exercise)

13
Log File - Example
14
Implementation

• Using event based simulation principles
• Main modules
• Simulator
• Maintains a priority queue of events
• Types of events job arrival, job termination
• Scheduler
• Maintains status of currently running jobs
• Maintains the wait queue
• Schedules jobs instructs the simulator when to
  start

15
Scheduler Implementation Issues

• Two types of schedulers
• Consider common and distinct features
• Simulator should work with different schedulers
• Note scheduler can instruct initiation of
  several processes simultaneously
• See Strategy pattern in design patterns

16
Parsing Log Files

• Scanner is too slow for reading complete large
  files
• Solution
• Use BufferedReader and FileReader
• Read each line using Scanner
• Integers can be read using nextInt method of
  Scanner

17
Implementing a Priority Queue

• Can be implemented using a heap (required in the
  exercise)
• Assume heap size is not known in advance
• Handle memory allocation carefully
• Initialize an array with a reasonable size
• Inserting a new element to a full heap
• How do we expand the heap?