GridSim 2.0 Adv. Grid Modelling

1
GridSim 2.0Adv. Grid Modelling Simulation
Toolkit

• Rajkumar Buyya, Manzur Murshed (Monash),
• Anthony Sulistio, Chee Shin Yeo

Grid Computing and Distributed Systems (GRIDS)
Lab, Dept. of Computer Science and Software
Engineering The University of Melbournewww.gridbu
s.org
Thanks to David Abramson
2
Outline

• Motivation.
• System Architecture.
• GridSim Entities.
• Visual Modeller.
• Experiments.

3
Performance Evaluation With Large Scenarios

• Varying the number of
• Resources (1 to 100s..1000s..).
• Resource capability.
• Cost (Access Price).
• Users.
• Deadline and Budget.
• Workload.
• Different Time (Peak and Off-Peak).
• We need a repeatable and controllable
  environment.
• Can this be achieved on Real Grid testbed ?

4
Grid Environment

• Dynamic
• Resource and User Properties vary with time.
• Experiment cannot be repeated.
• Resources are distributed and owned by different
  organisations. Heterogeneous users.
• It is hard to create a controllable environment.
• Grid testbed size is limited.
• Also, creating testbed infrastructure is time
  consuming and expensive.
• Hence, grid computing researchers turn to
  modelling and simulation.

5
GridSim Toolkit

• GridSim 1.0 released in Dec. 2001
• GridSim and GridBroker.
• GridSim 2.0 released in Nov. 2002 _at_ SC 2002.
• Improvements in GridSim and GridBroker.
• Add Visual Modeler.
• Few functionalities of GridSim
• Allows modelling of heterogeneous of resources
  users.
• Supports simulation of both static dynamic
  schedulers.
• Simulates applications with different parallel
  models.

6
System Architecture
Application, User, Grid Scenarios Input and
Results
. . .
Application Configuration
Resource Configuration
Visual Modeler
Grid Scenario
Output
Grid Resource Brokers or Schedulers
GridSim Toolkit
Application Modeling
Information Services
Resource Allocation
Statistics
Job Management
Resource Entities
Resource Modeling and Simulation (with Time and
Space shared schedulers)
Clusters
Single CPU
Reservation
SMPs
Load Pattern
Network
Basic Discrete Event Simulation Infrastructure
SimJava
Distributed SimJava
Virtual Machine (Java, cJVM, RMI)
Distributed Resources
PCs
Workstations
Clusters
SMPs
7
GridSim Entities
ShutdownSignal Manager i
Broker i
Internet
User i
Output
Output
Resource j
Scheduler
Application
Job Out Queue
Jobs
Process Queue
Job In Queue
Input
Input
Input
Output
Statistics Recorder i
Report Writer i
Resource List
InformationService
8
GridSim Entities Communication Model
9
Time Shared Multitasking and Multiprocessing
Tasks onPEs/CPUs
P1-G2
P1-G1
P3-G2
P1-G3
P2-G3
P2-G2
G3
G2
G3
PE2
G2
PE1
G1
G2
2
6
9
12
16
19
26
22
Time
G1
G1F
G3
G2
G2F
G3F
G1
G1 Gridlet1 Arrives
Gridlet1 (10 MIs)
G1F Gridlet1 Finishes
G2
Gridlet2 (8.5 MIs)
P1-G2 Gridlet2 didnt finish at the 1st
prediction time.
G3
P2-G2 Gridlet2 finishes at the 2nd prediction
time.
Gridlet3 (9.5 MIs)
10
Space Shared Multicomputing
Tasks onPEs/CPUs
P1-G1
P1-G2
P1-G3
G2
G3
PE2
G1
G3
PE1
2
6
9
12
16
19
26
22
Time
G1
G1F
G3
G2
G2F
G3F
G1
G1 Gridlet1 Arrives
Gridlet1 (10 MIs)
G1F Gridlet1 Finishes
G2
Gridlet2 (8.5 MIs)
P1-G2 Gridlet2 finishes as per the 1st
Predication
G3
Gridlet3 (9.5 MIs)
11
Visual Modeler

• Available in GridSim 2.0
• Functionalities
• Create and delete many users and resources.
• Able to save and load the model file (XML
  format).
• Generate Java source code.

12
Experiment 1

• Create 21 users and 25 resources.
• Cost varies from 10 to 20 units per sec (G/sec).
• Each user has 20 jobs with variation of 2.
• Want to optimise cost.
• Simulation Time approx. 7 hours.
• Number of users grows -gt Pr (one resource per
  user) decreases.
• This low Pr demands high D_Factor and B_Factor in
  order to achieve very high job completion rate.

13
D-Factor

• Any job with D_Factor lt 0 would never be
  completed.
• As long as some resources are available
  throughout the deadline, any job with D_Factor ?
  1 would always be completed.

14
B-Factor

• Any job with B_Factor lt 0 would never be
  completed.
• As long as some resources are available
  throughout the deadline, any job with B_Factor ?
  1 would always be completed.

15
Main Window of Visual Modeler
16
View User Property Dialog
17
View Resource Property Dialog
18
Job Completion Cost Optimise
19
Time Utilisation Cost Optimise
20
Budget Utilisation Cost Optimise
21
Experiment 2

• Workload Synthesis
• 200 jobs, each job processing requirement 10K
  MI or SPEC with random variation from 0-10.
• Exploration of many scenarios
• Deadline 100 to 3600 simulation time, step
  500.
• Budget 500 to 22000 G, step 1000.
• Deadline and Budget Constraint (DBC) Strategies
  
• Cost Optimisation for a single user.
• Resources Simulated WWG resources.

22
Simulated WWG Resources
23
Gridlets vs Budget
24
Impact of budget for deadline values
25
Budget spent with deadline values
26
Selected GridSim Users
27
Conclusion

• GridSim toolkit is suitable for application
  scheduling simulations in Grid and P2P computing
  environment.
• GridSim 2.0 is available to download
• www.gridbus.org
• Extending to support Data Grid modelling