Constructing A Grid Simulation with
Differentiated Network Service
using GridSim
Anthony Sulistio, Gokul Poduval, Rajkumar Buyya, Chen-Kong Tham
Fellow of Grid Computing
Grid Computing and Distributed Systems (GRIDS) Lab.
The University of Melbourne, Australia
Networks and Distributed Systems Lab
National University of Singapore (NUS), Singapore.
www.gridbus.org/gridsim/
1
Presentation Outline
•
•
•
•
•
•
•
Introduction
Background
Design and Implementation
Experiments and Results
Related Work
Conclusion and Further Work
Questions and Answers
2
Grid as Cyberinfrastructure for
e-Science and e-Business Applications
Grid Information Service
Grid Resource Broker
R2
Application
database
R3
R5
R4
RN
Grid Resource Broker
R6
R1
Resource Broker
Grid Information Service
3
Resource Management and
Application Scheduling

This is one of most challenging aspect of
Grid Computing:


Due to presence of heterogeneity resources
along dynamic variation of available
capability of resources.
Application Scheduling Policies need to
properly investigated/evaluated before
deploying them on production Grids.
4
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 ?
5
Grid Environment
• Dynamic:
1. Resource and User Properties vary with time.
 Experiment cannot be repeated.
2. Resources are distributed and owned by
different organizations. 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
modeling and simulation.
6
GridSim Toolkit


GridSim is a Java-based discrete-event grid
simulation package.
GridSim is based on SimJava2.
• Few functionalities of GridSim:
 Allows modeling of heterogeneous of various types of
resources & users.
 Resources can be extended to implement your own
allocation policies (e.g, SLA or VO based allocation).
 Supports simulation of both static & dynamic schedulers.
 Simulates applications with different parallel models.
7
GridSim - System Architecture
Application, User, Grid Scenario’s Input and Results
Application
Configuration
Resource
Configuration
Visual
Modeler
Grid
Scenario
...
Output
Grid Resource Brokers or Schedulers’s Simulation
GridSim Toolkit
Application
Modeling
Resource
Entities
Information
Services
Job
Management
Resource
Allocation
Statistics
Resource Modeling and Simulation (with Time and Space shared schedulers)
Single CPU
SMPs
Clusters
Load Pattern
Network
Add your own policy
for resource allocation
Reservation
Basic Discrete Event Simulation Infrastructure
SimJava
Distributed SimJava
Virtual Machine (Java, cJVM, RMI)
PCs
Workstations
SMPs
Clusters
Distributed Resources
8
Network Functionalities



Communication networks serve as a
fundamental component of grid computing.
A realistic simulation of grid environments
should include the effects of sending data
over shared communication lines.
Earlier versions of GridSim did not have the
ability to specify a network topology, nor the
functionality to connect resources through
network links in the experiment.
9
Our Work

In this work, GridSim has been extended to
address the above problems with the
ability to simulate realistic network models
by:




allowing users to create a network topology,
packetizing a data into smaller chunks for
sending it over a network,
generating background traffic, and
incorporating different level of services for
sending packets.
10
GridSim Entities Communication Model
EA
EB
body()
body()
Send(output, data, EB)
…
…
Receive(input, data, EA)
…
…
Input_EA
body()
Input_EB
…
body()
…
Output_EA
Output_EB
body()
body()
…
…
data, t2
Timed Event Delivery
11
New Network Extension Model
New functionalities:
• support for Network Quality of Service, such as each packet has
a Type of Service (ToS) attribute
• support for Runtime Information, such as an ICMP ping
message.
• generate background traffic, which is done by Output
12
Experiment


The main aim of this experiment is to show
GridSim's ability to simulate an adequate-size grid
testbed.
For this experiment, we are mainly concern about
the network behavior in a grid environment. Hence,
we are trying to look at:


how background traffic might affect network loads and
overall execution time; and
how differentiated QoS for packets might help in a heavy
load situation;
13
Australian BADG test-bed – Hardware

Uni.Adelaide CS group


APAC/GrangeNet (at ANU)


1 P4 Intel 2.0GHz
70 GB disk
Uni.Melbourne GridBus/CS


2 Xeon 2.6GHz (IBM)
70 GB disk
Uni.Melbourne EPP group


2 Xeon 2.6GHz (IBM)
70 GB disk
2 Xeon 2.6GHz (IBM)
70 GB disk
Uni.Sydney HEP group

2 Xeon 2.6GHz (IBM)
70 GB disk
14
Experiment Setup



Five resources are created in four different
locations: Canberra, Adelaide, Melbourne
and Sydney.
All resources are connected via GrangeNet,
a Gigabit wide-area network within Australia.
All links share same characteristics, i.e. MTU
size of 1,500 bytes and latency of 10
milliseconds.
15
GrangeNet and Grid Modeling
16
Resource Characteristics
Name
Location
Resource Characteristics
Num
CPU
A SPEC
Rating
R0
Dept. of Physics,
Univ. of Melbourne
PC with Intel Pentium 2.0 Ghz,
512MB RAM
1
684
R1
GRIDS Lab,
Univ. of Melbourne
Dual Intel Xeon 2.6 Ghz, 2GB RAM
4
1050
R2
Dept. of Physics,
Univ. of Sydney
Dual Intel Xeon 2.6 Ghz, 2GB RAM
4
1050
R3
Dept. of Computer Sc.,
Univ. of Adelaide
Dual Intel Xeon 2.6 Ghz, 2GB RAM
4
1050
R4
Australia National Univ.,
Canberra
Dual Intel Xeon 2.6 Ghz, 2GB RAM
4
1050
Table 1.
Australian Belle analysis data grid testbed simulated using GridSim
17
User Characteristics

There are 5 users located on each of the four
locations, sharing the same characteristics:







bandwidth: 100 Mbps connected to a leaf router of each
testbed site
total number of jobs: 20 each
job data size: 1 MB each
job processing power: 100 Million Instructions (MI) each
job submission: uniformly distributed among five resources
as mentioned in Table 1.
background traffic: submits to all resources and other users,
with inter-arrival time using a Poisson distribution approach
with mean of 5 minutes.
Total number of packets for each interval is uniformly
distributed in
[1 ... 10]. The size of each packet is 1,500
bytes
18
Results: Advantage of network QoS in
a shared network environment
Priority
With background traffic
(in simulation minutes)
High
22.82
Normal
23.57
Table 2. Network QoS using SCFQ (self clocked fair queuing) packet scheduler
(4 users out of 20 are given high priority for sending their jobs)
Priority
With SCFQ scheduler
(in simulation seconds)
High
1.20 x 10-6
Normal
2.38 x 10-6
Table 3. An Average Packet Lifetime at the Melbourne Leaf Router
(which links 2 resources, hence more traffic that other leaf routers)
19
Results
Packet Scheduling
With background
traffic (seconds)
Without background
traffic (seconds)
SCFQ
122 x 10-6
121 x 10-6
FIFO
149 x 10-6
146 x 10-6
Table 4.
An average of high priority package
lifetime at the Melbourne Leaf Router
under a heavier load (job data size =
10MB, previously 1 MB)
20
Results: effect of background traffic
Number of packets passing through the Melbourne Leaf Router
21
Related Work
Simulation
Tools
Routing
Table
Entry
Type of Transport
Protocol
Data
Packetization
Runtime
Network
Status
Network
QoS
GridSim
Automatic
A datagram oriented
protocol similar to UDP
Supported
Supported
Supported
MicroGrid
Automatic
TCP and UDP
Supported
Supported
Not
supported
SimGrid
Manual
TCP
Not supported
Supported
Not
supported
OptorSim
Manual
Not supported
Not supported
Not
supported
Not
supported
Table 5. Listing of network functionalities and features for
each grid simulator
22
Conclusion
• GridSim toolkit provides comprehensive support
application scheduling simulations in Grid
computing environment.
• GridSim has new features such as generating
background traffic during an experiment, requesting
network information during runtime and providing
differentiated service for packets based on users‘
Quality of Service (QoS) requirements.
• Our experiment has shown how GridSim can be
used to simulate a medium-sized grid testbed.
• GridSim is available to download:
www.gridbus.org/gridsim/
23
Future Work

We are planning to incorporate additional features
into GridSim, such as




having different types of routing algorithms, schedulers
and reservation of network resources.
adding other type of network building blocks like
switches and domain gateways.
support will be added for non work-conserving routers.
planning an ability to design the network topology using
scripts similar to ns-2.
24
Selected GridSim Users
25