Docstoc

PowerPoint Presentation - Java CoG Kit

Document Sample
PowerPoint Presentation - Java CoG Kit Powered By Docstoc
					Grid Portal Development for
Sensing Data Retrieval and
Processing
Diego Arias, Mariana Mendoza, Fernando Cintron,
Kennie Cruz, and Wilson Rivera

Parallel and Distributed Computing Laboratory
University of Puerto Rico at Mayaguez

GCE06-SC06 – Tampa, FL
November 12-13, 2006
Agenda



 Introduction
 CASA: Collaborative and Adaptive Sensing of the Atmosphere
    – NSF: EEC-031374; UMass, UPRM, OU, CSU
 WALSAIP: Wide Area Large Scale Automated Information Processing
    – NSF: CISE-CNS-0424546; UPRM
 Remarks and conclusions




                               Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM   2
Introduction: The PDC Laboratory




                        Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM   3
Introduction: PDCLab Research


                    Optimized e-Resource
                                                                                                   PDC Portal
                       Administration


                   Automated Composition
                 Signal Processing Operators
                                                                                                   WALSAIP
                                                                                                    Portal
                 Adaptive Provisioning and
                Orchestration of Grid Services



                   Information Dispersal                                                           STB Portal



              Semi-Analytical Inversion Model                                                      CenSSIS
                                                                                                    Portal




                             Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM                4
Introduction: PDCLab Grid Testbed




                        Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM   5
Introduction: PDCLab Grid Testbed

 CentOS 4.2
 Globus Toolkit 4.0.1.
     –   OpenPBS
     –   Torque
     –   PosgreSQL
     –   Apache Ant version 1.6.5
     –   Java SDK version 1.5
     –   Jakarta Tomcat version 5.5.9.




                                         Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM   6
Introduction: PDCLab Grid Testbed




                        Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM   7
Agenda



 Introduction
 CASA: Collaborative and Adaptive Sensing of the Atmosphere
    – NSF: EEC-031374; UMass, UPRM, OU, CSU
 WALSAIP: Wide Area Large Scale Automated Information Processing
    – NSF: CISE-CNS-0424546; UPRM
 Remarks and conclusions




                               Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM   8
CASA: Collaborative and Adaptive Sensing




                                        10,000 ft




                         3.05 km




                                                                                                               4 km
                                                          1 km




                                                                                   2 km




                                                                                                                      5.4 km
                                                                               snow
                                                                                              wind

                                                                                                     tornado   gap
                        Horz. Scale: 1” = 50 km
                        Vert. Scale: 1” -=- 2 km
                                   0            40                80             120          160    200        240
                                                                               RANGE (km)




                                            gap - earth curvature prevents 72%
                                            of the troposphere below 1 km from
                                            being observed.
                        Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM                                    9
CASA: Student Testbed (STB)



                                                                                                               Government


                                                                                             Service/Request
                                                                                                   Data




                                                                                                                Media




                                                                                                               Academic Institutions/Researchers




                       Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM                                                10
CASA: STB Grid Portal Interface




                        Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM   11
CASA: STB Grid Portal Interface




                        Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM   12
CASA: Information Dispersal Algorithm




         Data File F




           IDA (12,4)



 1   3    9             10   6   4

                                                                                                                             Operation
                                                                                                             IDA - Dispersal Operation
 5   2    12            7    8   11
                                                                               1000
                                                                               1500
                                                                               1400
                                                                                900
                                                                               1300
                                                                                800
                                                                               1200
         Data Retrieval
                                                                               1100
                                                                                700
                                                          Elapsed Time (Sec)
                                                          Elapsed Time (Sec)




                                                                               1000                                                                                 IDA-OLD
                                                                                                                                                                      8MMA
                                                                               600
                                                                               900                                                                                  8MMA
                                                                                                                                                                      8MAP
                                                                               800                                                                                  8MAP
                                                                                                                                                                      8FRL
                                                                               500
                                                                               700                                                                                  8FRL
                                                                                                                                                                      16MMA
                                                                               400
                                                                               600                                                                                  16MMA
                                                                                                                                                                      16MAP
                                                                               500                                                                                  16MAP
                                                                                                                                                                      16FRL
                                                                               300
                                                                               400                                                                                  16FRL
                                                                               200
                                                                               300

                                                                               200
                                                                               100
                                                                               100
                                                                                 0
                                                                                 0
                                                                                      0
                                                                                      0   100
                                                                                          100   200
                                                                                                200    300
                                                                                                      300       400
                                                                                                               400     500
                                                                                                                      500     600
                                                                                                                             600     700
                                                                                                                                    700      800
                                                                                                                                           800       900
                                                                                                                                                   900       1000
                                                                                                                                                           1000
                                                                                                                 Data Size (MB)
                                                                                                                 Data Size (MB)


                                      Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM                                                                     13
CASA: Data Management Portlets




                       Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM   14
CASA: Weather Information Portlets




                        Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM   15
CASA: Job Submission




                                Resource Manager                                                                                Netcdf to JPG - CPU Usage

                                                                                                                  98   98   98              98
                                                                                                            100                     97                  97   97
                                                                                                                                                                  94   93   93
                                                                                                            90
                   JOB MANAGER                         Resources
                                                                                                            80
          Grid
         Portal                                                                                             70




                                                                                            CPU Usage (%)
                                           Process                                                          60
Client
                   GateKeeper              Process                                                          50
             GSI                                                                                            40
                                           Process
                                                                                                            30

                                                                                                            20

                                                                                                            10
                                                                                                                  2    2    1       1       1           1    1    0    0    0
                                                                                                             0
                                                                                                                  1    2    3        4          5        6   7    8    9    10

                                                                                                                                         Grid       Server




                                                     Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM                                                         16
Agenda



 Introduction
 CASA: Collaborative and Adaptive Sensing of the Atmosphere
    – NSF: EEC-031374; UMass, UPRM, OU, CSU
 WALSAIP: Wide Area Large Scale Automated Information Processing
    – NSF: CISE-CNS-0424546; UPRM
 Remarks and conclusions




                               Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM   17
WALSAIP: Signal Based Information Processing System


                        Physical                                               Observables
                                                                                                           Sensors
                       Environment                                                                         Effectors


                                                                                                       Signals


    Knowledge    Information                Information                                         Data         Signal
    Processing                              Processing                                                     Processing


          Knowledge


     Decision    Intelligence                             Physical
     System                                              Environment

                          Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM                           18
WALSAIP: Conceptual Framework Design


                                   RDS                               INTERNET                           CDS




                                                                                                                     IRS
         SAS       SCS         DRS
          SAS                                                              CSPS                        SDP           IRS
                                                                                                                     IRS
                                                                 Processing Stage
                                                                                                      Post-processing Stage
                  Pre-processing Stage

SAS: Sensor Array Structures                                                   SCS: Signal Conditioning System
DRS: Data Representation System                                                RDS: Raw Data Server
CSPS: Computational Signal Processing System                                   CDS: Computed Data Server
SDP: Signal Data Post-processing                                               IRS: Information Rendering System


                                Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM                           19
     CIP over a Cyberspace Infrastructure
     SAS      SCS         DRS                      RDS       DBMS




               CSPS                                       CSPS
                                     Net-Centric
                                       MIMO
                                      Services

    CDS                                                          IRS
               DBMS
                                                    SDP          IRS

                                           DRS                   IRS



             IRS
                                                   RDS       DBMS
             IRS       SDP                 SCS

             IRS
                                             SAS
CIP: Computational and Information
     Processing
     CIP over a Cyberspace Infrastructure
     SAS      SCS         DRS                      RDS       DBMS




               CSPS                                       CSPS
                                     Net-Centric
                                       MIMO
                                      Services

    CDS                                                          IRS
               DBMS
                                                    SDP          IRS

                                           DRS                   IRS



             IRS
                                                   RDS       DBMS
             IRS       SDP                 SCS

             IRS
                                             SAS
CIP: Computational and Information
     Processing
     CIP over a Cyberspace Infrastructure
     SAS      SCS         DRS                      RDS       DBMS




               CSPS                                       CSPS
                                     Net-Centric
                                       MIMO
                                      Services

    CDS                                                          IRS
               DBMS
                                                    SDP          IRS

                                           DRS                   IRS



             IRS
                                                   RDS       DBMS
             IRS       SDP                 SCS

             IRS
                                             SAS
CIP: Computational and Information
     Processing
     CIP over a Cyberspace Infrastructure
     SAS      SCS         DRS                      RDS       DBMS




               CSPS                                       CSPS
                                     Net-Centric
                                       MIMO
                                      Services

    CDS                                                          IRS
               DBMS
                                                    SDP          IRS

                                           DRS                   IRS



             IRS
                                                   RDS       DBMS
             IRS       SDP                 SCS

             IRS
                                             SAS
CIP: Computational and Information
     Processing
               WALSAIP’s Jobos Bay NERR Environmental Surveillance and Monitoring Testbed




                 @ Jobos Bay NERR                                                                                           @ UPRM


                                                                                   INTERNET
 MBR410
Base Station
                 RS-232
                                                                                                                   Remote User

                          Local Host


                                            Data Base
                                             on Host
                                                                                                                                       Sensing-Data

                                                                                                                       Database             SensorNode1
                                       MEP-410                                                                       Querying and
                                                                                                                      Processing
                                                                                                                        Engine              SensorNode5

                                                                                                                                                Cooked-data
RF-Link
                                                            MEP-510                                                                 1-Jun022006-21:02:00-humidity-data.txt

                                                                                                                                    1-Jun022006-21:02:00-humidity-meta.txt




                                             Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM                                                   24
WALSAIP: Grid Portal Interface




                        Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM   25
WALSAIP: Grid Portal Interface




                        Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM   26
WALSAIP: Grid Portal Interface




                        Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM   27
Agenda



 Introduction
 CASA: Collaborative and Adaptive Sensing of the Atmosphere
    – NSF: EEC-031374; UMass, UPRM, OU, CSU
 WALSAIP: Wide Area Large Scale Automated Information Processing
    – NSF: CISE-CNS-0424546; UPRM
 Remarks and conclusions




                               Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM   28
Remarks and Conclusions



 Grid portals for large scale applications
     – Sensing of the Atmosphere (CASA)
     – Environmental Surveillance and Monitoring (WALSAIP)
 Grid services provided via portlets
     –   Distributed data management
     –   Information dispersal
     –   Weather services
     –   Multiple job submission
     –   Service orchestration
 Future work
     – Remote instrumentation via grid portals
     – Integration to community grid infrastructures (Teragrid, etc….)




                                       Wilson Rivera, Parallel and Distributed Computing Laboratory @ UPRM   29

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:2
posted:2/25/2012
language:
pages:29