Taiwan UniGrid

1
Taiwan UniGrid

• Yeh-Ching Chung
• Department of Computer Science
• National Tsing Hua University
• Hsin-Chu, 300, Taiwan

2
Outline

• Introduction
• Portal
• Broker and Scheduler
• Resource Information Service
• Storage Service
• Applications
• Conclusion

3
Introduction (1)

• The purpose of grid computing is to integrate
  various resources within a large network
  environment.
• The purpose of the UniGrid project is to build a
  platform for academic research using grid-related
  technologies in Taiwan.

4
Introduction (2)

• 9 institutes join to develop the system
• ????
• ???????
• ??????
• ???????
• ???????
• ???????
• ???????
• ????????????
• ?????????

5
Introduction (3)

• All institutes that participate in the UniGrid
  project contribute some resources.
• These resources can be used in collaboration for
  large scale applications.

6
Introduction (4)

• System Architecture

7
Outline

• Introduction
• Portal
• Broker and Scheduler
• Resource Information Service
• Storage Service
• Applications
• Conclusion

8
Portal

• The UniGrid portal provides an interface for
  UniGrid users to use the resources available in
  the UniGrid system.
• Functionalities of the portal
• System status monitoring
• Single sign-on
• User workflow management
• Project information

9
System Status Monitoring (1)

• UniGrid users can examine the status of system
  resources through the portal.
• The portal gathers the current system information
  from the information service and present these
  information to the users.

10
System Status Monitoring (2)

• Screenshot of the system status monitoring web
  page

11
Single Sign-On (1)

• Single sign-on is a mechanism whereby a single
  authentication can permit a user to access all
  resources where he has access permission, without
  the need to enter multiple passwords.
• All user account information are kept in a
  database at the portal site.
• When a user requests a service, his verification
  data is passed to that service.
• The request will be granted only if the identity
  is verified by the verification web service

12
Single Sign-On (2)

• User identity verification through single sign-on
  service

13
User Workflow Management (1)

• A UniGrid user can design and save his own
  workflows at the UniGrid portal.
• A user can select any workflow he designed and
  execute the workflow through the UniGrid portal.
• A user can also monitor the status of his
  workflow through the UniGrid portal.

14
User Workflow Management (2)

• Structure of a workflow

Workflow
parallel execution
sequential execution
15
User Workflow Management (3)

• The workflows of each user is stored in the
  portal storage in XML format.
• ltflow name"testflow" numstages"3"gt
• ltstage name"stage1" numjobs"1"gt
• ltjob id"0"gt
• ltsortkeygt1lt/sortkeygt ltruntypegtmpilt/runtypegt
• ltworkdirgt/home/test/lt/workdirgt
• ltfilenamegtmm_mpilt/filenamegt
• ltrunrpgttruelt/runrpgt ltdatafile/gt ltargugt256lt/argugt
• ltotherurl/gt ltcpunogt4lt/cpunogt
• lt/jobgt
• lt/stagegt
• lt/flowgt

16
User Workflow Management (4)

• Screenshot of the workflow editing web page

17
User Workflow Management (5)

• When an user submits a workflow, the portal will
  pass the selected workflow information to the
  broker.
• Upon receiving an execution request, the resource
  broker will find the required resource for that
  workflow and schedule its execution.

18
User Workflow Management (6)
19
User Workflow Management (7)

• Users can examine the execution status of his
  workflow through the portals workflow monitoring
  system.
• All workflow execution information are stored in
  a database at the machine with resource broker
  installed on it.
• The portal queries the database and obtain the
  current status of a particular workflow.
• The status information is processed and presented
  in the form of web pages.

20
User Workflow Management (8)

• Screenshot of the workflow monitoring web page

21
User Workflow Management (9)

• Screenshot of the UniGrid workflow management web
  page

22
Outline

• Introduction
• Portal
• Broker and Scheduler
• Resource Information Service
• Storage Service
• Applications
• Conclusion

23
Broker Scheduler (1)

• The broker provides a uniform interface to access
  available resources in the UniGrid system.
• The broker uses the resource information service
  to obtain the current status of the resources in
  the system.
• After these information are gathered, the broker
  will allocate the resources that meets the
  requirements of the current job.
• The jobs are then passed to the corresponding
  local schedulers to be executed locally.

24
Broker Scheduler (2)

• Broker workflow

25
Broker Scheduler (3)

• Each participating organization has a local
  scheduler (Condor) installed to schedule the jobs
  assigned to that organization.
• Condor
• A scheduler for large collections of
  distributively owned computing resources
• Developed by the researchers at University of
  Wisconsin
• Specialized for compute-intensive jobs
• Uses the ClassAd mechanism to match job
  requirements to machine status and schedule the
  jobs according to the matching results

26
Related Research (1)

• Tools have been developed to simulate different
  load sharing and scheduling policies on computing
  grid and analyze their performance
• Queuing methods
• Independent clusters
• Multiple queues
• Forwarding to no-need-to-wait site
• Forwarding to shortest-queue site
• Forwarding to least-load site,
• load

27
Related Research (2)

• Queuing methods (contd.)
• Single queue
• Multi-pool centralized queue
• Single-pool centralized queue
• One big cluster
• Two-level scheduling
• Empty queue only
• Shortest queue first
• Least load first
• Two-level local queues
• Forwarding to shortest-queue site

28
Related Research (3)

• Scheduling policies
• Non-FCFS
• Multi-pool centralized queue
• Single-pool centralized queue
• FCFS
• Two-level scheduling
• The performance of Non-FCFS is three times better
  than FCFS

29
Related Research (4)

• Implementation Approaches
• Multi-Pool Centralized Queue
• Global queue scheduling in the broker, no local
  queuing system
• Global queue scheduling in the broker, making
  sure available processors through local queuing
  system
• Single-Pool Centralized Queue
• Global queue scheduling in the broker, no local
  queuing system

30
Related Research (5)

• Two-Level Scheduling (Empty-Queue-Only
  Multi-Pool Grid)
• Global queue in the broker, local queues in the
  local queuing systems

31
Related Research (6)

• Simulation results

32
Related Research (7)

• Simulation results (contd.)

33
Related Research (8)

• Discussion
• Non-FCFS methods can effectively improve the
  overall system utilization and performance.
• The smallest first non-FCFS policy outperforms
  all other policies in terms of waiting time and
  waiting ratio.
• As the worst case is concerned, the backfilling
  policy is superior because it does not allow jobs
  to be delayed by the backfilling activities

34
Outline

• Introduction
• Portal
• Broker Scheduler
• Resource Information Service
• Storage Service
• Applications
• Conclusion

35
Resource Information Services

• The resource information service provides
  information about current resource status, these
  information can be used by other services of the
  system
• Functionalities of the resource information
  service
• Information system
• Performance visualization of MPI parallel
  programs execution

36
Information System (1)

• Provides an interface for other services to query
  various information about computing nodes
• The statistics about the individual nodes are
  obtained using MDS (Monitoring Discovery
  Service) provided by the Globus Toolkit
• The current network status between machines are
  gathered using NWS (Network Weather Service)
• Automatic update of node information
• When a new computing nodes is added/removed

37
Information System (2)

• The Network Weather Service (NWS)
• A distributed system that periodically monitors
  and dynamically forecasts the performance various
  network and computational resources can deliver
  over a given time interval
• Developed by the researchers at UCSB
• It uses numerical models to generate forecasts of
  what the conditions will be for a given time
  frame
• Because this functionality is analogous to
  weather forecasting, the system is called Network
  Weather Service

38
Information System (3)
39
Information System (4)

• Screenshot of the node status webpage

40
Performance Visualization of MPI Programs (1)

• Input any application (depending on the
  availability of compiler in grid platform)
• Output performance visualization of the
  execution of this application

41
Performance Visualization of MPI Programs (2)

• Execution of a Parallel Application using 4
  computing nodes

42
Related Research (1)

• Communication localization data partitioning
  techniques in cluster-based grid system
• Localized communication enhances performance of
  parallel applications on grid
• Adaptive data partitioning for identical cluster
  non-identical cluster grid topology
• In-core out-of-core applications

43
Related Research (2)

• Communication localization techniques for
  identical cluster

Localized communication patterns
Original communication patterns
44
Related Research (3)

• Communication localization techniques for
  non-identical cluster

Original communication table
45
Related Research (4)

• Communication localization techniques for
  non-identical cluster (contd.)

Localized communication table
46
Outline

• Introduction
• Portal
• Broker and Scheduler
• Resource Information Service
• Storage Service
• Applications
• Conclusion

47
Storage Service

• The goal of storage service is to provide a
  collaborative space where UniGrid users can share
  their data and resources with others.
• Components of the storage service
• Virtual storage system
• Data management system

48
Virtual Storage System (1)

• Virtual storage system architecture

49
Virtual Storage System (2)

• The virtual storage system is implemented with
  Java as a web service
• UniGrid services access the virtual storage
  system when they need to fetch/modify users data
  files
• A client program is available for users to manage
  his own storage space
• The files are stored in a master file server and
  replicas of the files are distributed to other
  machines

50
Virtual Storage System (3)
51
Virtual Storage System (4)

• Screenshot of the storage service client program

52
Data Management (1)

• The Data Management is the Web-based Replica
  Access and Management System
• It consists of the Registration, Search and
  Manager system
• The registration system is used in managing the
  user for accessing the UniGrid System
• The search system combines with the RLS and Web
  technique
• The manager system offers a friendly interface
  for manager, it will be easy to maintain the
  contents of database

• Structure of the Data Management System

53
Data Management (2)

• Replica Location Service

54
Data Management (3)

• The Registration System
• In Security
• We design a web registration system
• User need to be registered in portal and logged
  in by CA (Proxy-init)
• In account manage
• Administrator
• User

• The detailed structure of Web Service System

55
Data Management (4)

• The Search System
• Replica Index and Replica Location
• In LRC Sever, we can execute the basic command.
• We can update information of LRI server use the
  batch command
• Services
• We offer the service of the Job submit, files
  list, files upload and data replication in single
  server

• The detailed structure of Web Service System

56
Data Management (5)

• The Manager System
• We plan to design a friendly interface for
  manager, it will be easy to maintain the contents
  of Metadata database, update the RLS database and
  manage users account

• The detailed structure of Web Service System

57
Outline

• Introduction
• Portal
• Broker and Scheduler
• Resource Information Service
• Storage Service
• Applications
• Conclusion

58
Applications 1
Simulations of atmospheric circulations with the
NTU/Purdue nonhydrostatic numerical model. Model
characteristics Nonhydrostatic Explicit
forward-backward integration for both
high-frequency waves and gravity waves Implicit
diffusion scheme with a TKE prognostic
equation Time split schemes for high-frequency
waves, gravity waves, diffusion, and surface
processes. Physical processes Cloud
microphysics Surface similarity equation 3-layer
soil model Coriolis force
59
Performance with the UniGrid
12 hr
50 sec
30 min
5 hr
17 min
12 hr
50 sec
31 min
5 hr
35 min
60
Commands for submitting jobs
/opt/mpich/pgi/bin/mpirun nolocal machinefile
host np 8 nonh3d.exe gt test
host
NTU uninode11 2 uninode12 2 uninode14
2 uninode15 2
makefile
OBJ nonh3d.o tograds.o copy.o update.o sound.o
adv.o cloud1.o dampini.o\ initial.o
restart.o nbr2d.o startend.o tkeeq.o updtrp.o\
sprogi4.o sprogi2.o diffxy.o diffz.o pbl.o EXE
../nonh3d.exe OPT -O3 -Mextend -Msave
-Bstatic -byteswapio OPT -O3 -static
-ffixed-line-length-80 OPT -O3 -static OPT
-O3 OPT -static (EXE) (OBJ)
/opt/mpich/pgi/bin/mpif77 (OPT) -o (EXE)
(OBJ) .f.o /opt/mpich/pgi/bin/mpif77
(OPT) -c lt clean rm -f .o
../nonh3d.exe
host1
uninode11 uninode11 uninode10 uninode10 uninode12
uninode12 uninode14 uninode14 uninode15 uninode15
uninode5 uninode5 uninode7 uninode7 uninode9 unino
de9
61
Three-dimension simulation of a thermal bubble
in an isentropic environment
Initial spherical bubble develops into a
mushroom-like shape. Two isentropic surfaces are
shown. The isentropic surface corresponding to a
higher potential temperature is in pink.
62
Two-dimensional simulation of a sea breeze
10 5km 0
z
SBF
0
15km 30
x
The figure shows the total water mixing ratio
(vapor plus liquid) over land after 2.5 hr. The
label under the x-axis is the distance from the
coastline. Water vapor is pumped up from the
ground surface in the convective boundary layer
(with the red/orange color representing high
water vapor content in the air). The location of
the sea breeze front (SBF) is shown.
63
Applications 2

• FASTA
• Compares a protein sequence to another protein
  sequence or to a protein database, or a DNA
  sequence to another DNA sequence or a DNA library

64
Applications 3

• ClustalW
• A general purpose multiple sequence alignment
  program for DNA or proteins.

65
Conclusions and Future Work

• A prototype of UniGrid system has been developed
• Enhance the data grid part of UniGrid
• Promote the UniGrid system to universities in
  Taiwan