1. CSS490 Grid Computing Textbook No Corresponding Chapter
Instructor: Munehiro Fukuda
These slides were compiled from The Grid: Blueprint for a New Computer Infrastructure.
Hello! Everyone,
My name is Shinya Kobayashi. Today, I am going to present our paper titled “Inter-Cluster Job Coordination Using Mobile Agents” on behalf of the first author, Munehiro Fukuda. Munehiro was hoping to show up and present the paper at AMS2001, however he got to wait in Japan until he will get an H1B visa. Since I received the presentation materials from him quite recently, please allow me to present this paper using this script.
I can respond to your questions as far as I know, however you can also ask Munehiro by . His address is on the title page of our paper.
(time 1:05)
Winter, 2004
CSS490 Grid Computing

2. Network Infrastructure
Users login their organizational systems first locally or remotely.
If they are affiliated with other organizations,
They can login from the system of their main use to some other systems. (They are given an opportunity to use those resources in parallel).
Problems:
They must orchestrate job execution among the resources they use.
Should those resources be limited to such a handful number of researchers?
High-speed
Information high way
Winter, 2004
CSS490 Grid Computing

3. The Computational Grid
Use computing resource connected to high-speed information highway as if we use electric power grid
Only 30% utilization in academic/commercial environments.
Many applications have only episodic requirements. So, why don’t we share computation resource?
Computational results and data should be also made available to all users.
Users:
Computational scientists and engineers
Experimental scientists
Association and corporations
Training and education
Consumers (E-commerce)
Winter, 2004
CSS490 Grid Computing

4. Grid Applications Category Examples Characteristics
Distributed supercomputing
DIS and Stellar dynamics
Very large problems needing lots of computing resource at a time
High throughput
Chip design and parameter studies
Harnessing many idle resources to increase aggregate throughput
On demand
Medical instrumentation
Allocating special resource dynamically
Data intensive
Sky survey
Using distributed data and needing high-volume data flows
Collaborative
Collaborative design
Education
Support communication or collaborative work
Winter, 2004
CSS490 Grid Computing

5. Example: Distributed Interactive Simulation
Fighter simulator
Tank simulator
Observer/Analyst
Small unit (company, platoon) commander
Software agents (automated enemy)
Winter, 2004
CSS490 Grid Computing

6. Grid Services Architecture (from www.globus.org slide)
High-energy
physics data
analysis
Collaborative
engineering
On-line
instrumentation
Applications
Regional
climate studies
Parameter
studies
Distributed
computing
Collab.
design
Remote
control
Application
Toolkit Layer
Data-
intensive
Remote
viz
Grid Services
Layer
Information
Resource mgmt
. . .
Security
Data access
Fault detection
Transport
. . .
Multicast
Grid Fabric
Layer
Instrumentation
Control interfaces
QoS mechanisms
Winter, 2004
CSS490 Grid Computing

7. Grid Systems/Tools NetSolve Legion Globus Toolkit
RPC-based approach
Legion
Object-based approach
Globus Toolkit
Tool-based approach
Winter, 2004
CSS490 Grid Computing

8. NetSolve RPC-based approach Clients Agents Servers
Network of servers
Client
RPC-based approach
Clients
Include a set of APIs called as (asynchronous) RPCs
Agents
Match client’s requests for services with servers
Servers
Encapsulates remotely accessed numerical libraries
Agent
Agent
choice
Scalar
server
Client
request
reply
MPP servers
Winter, 2004
CSS490 Grid Computing

9. Legion Legion classes Core objects Per-Program Scheduling
Act as managers and make policy
Core objects
Provide mechanisms that classes use to implement policies: hosts (processors), vaults(memory), context, binding agents, etc.
Per-Program Scheduling
Participating sites can assure their local policies.
User can choose a scheduling policy.
Prog
request
Enactor
Scheduler
Converted Legion object ID
By context objects
reserve
search
Converted Logion object address
By binding agents
Resource
database
Class
Host
collection
tty
Host
Host
tty
Resources
Class
tty
Winter, 2004
CSS490 Grid Computing

10. Globus A collection of tools
Resource brokers and resource co-allocators
MPI, HPC++, CORBA
Arch. Type, OS, memory, bandwidth
GRAM
(RSL-based management)
Nexus
(Remote execution by thread migration)
MDS
(Directory info. Tree)
Condor
NQE
Load Sharing Facility
IP, Message passing, ATM, shared mem.
LDAP
A collection of tools
GRAM: Resource allocation and process management
Nexus: Communication services
MDS: Directory services
GSI: Authentication/security services
Winter, 2004
CSS490 Grid Computing

11. CoordAgents (The CSS490 Instructor’s New Research Project)
Mobile-agent-based approach
Resources:
Server resources and client requests are all described in XML.
Servlet:
Launches a mobile agent on behalf of a client.
Maintains a server resources.
Mobile agents:
Carries a client program, executes it remotely, and reports its results to the client.
Each Server
Client
Outgoing
Agent
Web
Launcher
servlet
Incoming
Agent
Rsc Mgmt
servlet
Jave DB
Agent
engine
Resource
XML OS
Winter, 2004
CSS490 Grid Computing

12. CoordAgents Cont’d (The CSS490 Instructor’s New Research Project)
job
idle
powered
off!!
checkpointing
migration
PCs
Most PCs are idle
PC Grid
Why don’t PC users get together to use each other’s idle PCs?
Challenges
PCs may be powered on and off suddenly.
Network bandwidth may change suddenly.
Using CoordAgents
Dispatches a user job.
Performs check-pointing.
Migrate a process dynamically.
Winter, 2004
CSS490 Grid Computing