Presentation is loading. Please wait.

Presentation is loading. Please wait.

PRAGMA18 Demonstration 2-4 March 2010 Kei Kokubo, Yasuyuki Kusumoto, Susumu Date Osaka University, Cybermedia Center Wen-Wai Yim, Jason Haga Department.

Published byRalf Lucas Modified over 8 years ago

Similar presentations

Presentation on theme: "PRAGMA18 Demonstration 2-4 March 2010 Kei Kokubo, Yasuyuki Kusumoto, Susumu Date Osaka University, Cybermedia Center Wen-Wai Yim, Jason Haga Department."— Presentation transcript:

1 PRAGMA18 Demonstration 2-4 March 2010 Kei Kokubo, Yasuyuki Kusumoto, Susumu Date Osaka University, Cybermedia Center Wen-Wai Yim, Jason Haga Department of Bioengineering, University of California, San Diego

2 Overlay Network Organization A Organization B Organization C Physical network VM Virtual Cluster  The concept:  Dynamically aggregate available resources from different organizations on the basis of user’s request  Provide a user with a private virtual cluster Virtual node

3 1. Utilize three nodes from each of two clusters located at Osaka University 2. Flow of Demonstration 1. access to a Web portal to build a virtual cluster 2. set and submit some requirements for the deployment 3. submit an example job into the virtual cluster Overlay Network mocalatte VM Web Portal Virtual Cluster

4

5

6

7

8  Xen  Need for virtualization  OpenVPN  Need to provide a single private network address space among virtual machines and user’s machine  PIAX  Need to organize an overlay network  Discover appropriate resources for a user  Realize IP-based end-to-end communication on the overlay network relaying the connection of the OpenVPN  Java 5 or latter versions are needed

9

10

11  We believe our system will help scientists and researchers to deploy their private virtual cluster on available resources from different organizations.  Wen-wai actually verified the possibility of our proposed virtual cluster  The virtual cluster deploys an appropriate computational environment for her application.

12 Developed by UCSF Free, constantly improving Models ligand/receptor interactions – Dock score: sum of the van der Waals attractive, van der Waal dispersive and Coulombic electrostatic energies – Time – http://dock.compbio.ucsf.edu/Overview_of_DOCK/index.htm Ligands/Receptor from ZINC database – http://zinc.docking.org

13 Developer's nacona komolongm aocikbpraauroracafe01tea01 OS-bit6432 machine bit6432 gcc version 3.4.64.1.23.4.6 IDScores ZINC0001356410.801939-9.4002210.801939-9.40021 ZINC0015086321.139143-12.467221.13914321.13914 ZINC0015226530.23836130.2380330.23836119.05393 ZINC0015844252.5658852.5657952.56588 ZINC00158751-1.058107-1.05816-1.058107-1.05811 ZINC01555236503.2497255.02E+08503.249725 Graph I. Visualization of DOCK 6.2 scores over PRAGMA Grid Graph I. DOCK 6.2 scores from virtual screenings across PRAGMA clusters Developer's naconakomolongmaocikbpraauroracafe01tea01 OS-bit6432 machine bit6432 gcc version 3.4.64.1.23.4.6 IDScores ZINC0001356410.801939-9.4002210.801939-9.40021 ZINC0015086321.139143-12.467221.13914321.13914 ZINC0015226530.23836130.2380330.23836119.05393 ZINC0015844252.5658852.5657952.56588 ZINC00158751-1.058107-1.05816-1.058107-1.05811 ZINC01555236503.2497255.02E+08503.249725 Developer's naconakomolongmaocikbpraauroracafe01tea01 OS-bit6432 machine bit6432 gcc version 3.4.64.1.23.4.6 IDScores ZINC0001356410.801939-9.4002210.801939-9.40021 ZINC0015086321.139143-12.467221.13914321.13914 ZINC0015226530.23836130.2380330.23836119.05393 ZINC0015844252.5658852.5657952.56588 ZINC00158751-1.058107-1.05816-1.058107-1.05811 ZINC01555236503.2497255.02E+08503.249725

14 Developer'scoregridLatteVC_coregridVC_latte OS-bit6432 machine bit64326432 gcc version4.1.2 3.4.6 IDScores ZINC0001356410.801939-9.4002210.80194-9.40021 ZINC0015086321.139143-12.467221.13914 ZINC0015226530.23836130.2380330.2383619.05393 ZINC0015844252.5658852.5657952.56588 ZINC00158751-1.058107-1.05816-1.05811 ZINC01555236503.2497255.02E+08503.2497 coregridlatteVC_coregridVC_latte Time Per Molecule13.5 ± 0.448 ± 0.418.8 ± 0.352 ± 7 Total674.6 ± 0.52420 ± 5953 ± 22460 ± 6 Table II. DOCK 6.2 Scores of Virtual Clusters Compared with Physical Clusters Table III. DOCK 6.2 Execution Times on Virtual Clusters Compared with Physical Clusters Developer'scoregridLatteVC_coregridVC_latte OS-bit6432 machine bit64326432 gcc version4.1.2 3.4.6 IDScores ZINC0001356410.801939-9.4002210.80194-9.40021 ZINC0015086321.139143-12.467221.13914 ZINC0015226530.23836130.2380330.2383619.05393 ZINC0015844252.5658852.5657952.56588 ZINC00158751-1.058107-1.05816-1.05811 ZINC01555236503.2497255.02E+08503.2497 Developer'scoregridLatteVC_coregridVC_latte OS-bit6432 machine bit64326432 gcc version4.1.2 3.4.6 IDScores ZINC0001356410.801939-9.4002210.80194-9.40021 ZINC0015086321.139143-12.467221.13914 ZINC0015226530.23836130.2380330.2383619.05393 ZINC0015844252.5658852.5657952.56588 ZINC00158751-1.058107-1.05816-1.05811 ZINC01555236503.2497255.02E+08503.2497 Developer'scoregridLatteVC_coregridVC_latte OS-bit6432 machine bit64326432 gcc version4.1.2 3.4.6 IDScores ZINC0001356410.801939-9.4002210.80194-9.40021 ZINC0015086321.139143-12.467221.13914 ZINC0015226530.23836130.2380330.2383619.05393 ZINC0015844252.5658852.5657952.56588 ZINC00158751-1.058107-1.05816-1.05811 ZINC01555236503.2497255.02E+08503.2497 Developer'scoregridLatteVC_coregridVC_latte OS-bit6432 machine bit64326432 gcc version4.1.2 3.4.6 IDScores ZINC0001356410.801939-9.4002210.80194-9.40021 ZINC0015086321.139143-12.467221.13914 ZINC0015226530.23836130.2380330.2383619.05393 ZINC0015844252.5658852.5657952.56588 ZINC00158751-1.058107-1.05816-1.05811 ZINC01555236503.2497255.02E+08503.2497 Developer'scoregridLatteVC_coregridVC_latte OS-bit6432 machine bit64326432 gcc version4.1.2 3.4.6 IDScores ZINC0001356410.801939-9.4002210.80194-9.40021 ZINC0015086321.139143-12.467221.13914 ZINC0015226530.23836130.2380330.2383619.05393 ZINC0015844252.5658852.5657952.56588 ZINC00158751-1.058107-1.05816-1.05811 ZINC01555236503.2497255.02E+08503.2497

15

16  Performing DOCK 6.2 experiments on a heterogeneous Grid can lead to inconsistent results.  For any scientific application, this can be a problem.  Virtual clusters can provide a solution with affordable overhead costs.

17 Overlay Network Organization A Organization B Organization C Physical network VM Virtual Cluster  Develops P2P overlay network on every resources  Dynamically gathers unused computational resources from different organizations  Forms a virtual private network among the virtual machines Virtual node

18 1. Utilize computational resources of two clusters located at Osaka University 2. access to a Web portal to develop a virtual cluster 3. submit some requirement for the development Overlay Network mocalatte VM Virtual Cluster Web Portal

Download ppt "PRAGMA18 Demonstration 2-4 March 2010 Kei Kokubo, Yasuyuki Kusumoto, Susumu Date Osaka University, Cybermedia Center Wen-Wai Yim, Jason Haga Department."

Similar presentations

A Communication Relay Mechanism toward Construction of Virtual Cluster on Orverlay Network PRAGMA14 Working Group 11-12 March 2008 Yasuyuki Kusumoto Osaka.

A Communication Relay Mechanism toward Construction of Virtual Cluster on Orverlay Network PRAGMA14 Working Group March 2008 Yasuyuki Kusumoto Osaka.
Virtual Screening for SHP-2 Specific Inhibitors Using Grid Computing By Simon Han UCSD Bioengineering 09 November 18-21, 2008 SC08, Austin, TX.

Virtual Screening for SHP-2 Specific Inhibitors Using Grid Computing By Simon Han UCSD Bioengineering 09 November 18-21, 2008 SC08, Austin, TX.
PRAGMA Futures Panel Philip Papadopoulos, UCSD All Opinions are mine and do not necessarily reflect those of my university, the US government or my cat.

PRAGMA Futures Panel Philip Papadopoulos, UCSD All Opinions are mine and do not necessarily reflect those of my university, the US government or my cat.
A Workflow Engine with Multi-Level Parallelism Supports Qifeng Huang and Yan Huang School of Computer Science Cardiff University 2005.9.

A Workflow Engine with Multi-Level Parallelism Supports Qifeng Huang and Yan Huang School of Computer Science Cardiff University
Ravi Sankar Technology Evangelist | Microsoft

Ravi Sankar Technology Evangelist | Microsoft
High Performance Computing Course Notes 2007-2008 Grid Computing.

High Performance Computing Course Notes Grid Computing.
An OpenFlow based virtual network environment for Pragma Cloud virtual clusters Kohei Ichikawa, Taiki Tada, Susumu Date, Shinji Shimojo (Osaka U.), Yoshio.

An OpenFlow based virtual network environment for Pragma Cloud virtual clusters Kohei Ichikawa, Taiki Tada, Susumu Date, Shinji Shimojo (Osaka U.), Yoshio.
Dinker Batra CLUSTERING Categories of Clusters. Dinker Batra Introduction A computer cluster is a group of linked computers, working together closely.

Dinker Batra CLUSTERING Categories of Clusters. Dinker Batra Introduction A computer cluster is a group of linked computers, working together closely.
The ATHA Environment: Experience with a User Friendly Environment for Opportunistic Computing M.A.R.Dantas Department of Informatics (INE) University of.

The ATHA Environment: Experience with a User Friendly Environment for Opportunistic Computing M.A.R.Dantas Department of Informatics (INE) University of.
6th Biennial Ptolemy Miniconference Berkeley, CA May 12, 2005 Distributed Computing in Kepler Ilkay Altintas Lead, Scientific Workflow Automation Technologies.

6th Biennial Ptolemy Miniconference Berkeley, CA May 12, 2005 Distributed Computing in Kepler Ilkay Altintas Lead, Scientific Workflow Automation Technologies.
6/2/20071 Grid Computing Sun Grid Engine (SGE) Manoj Katwal.

6/2/20071 Grid Computing Sun Grid Engine (SGE) Manoj Katwal.
IT:Network:Applications VIRTUAL DESKTOP INFRASTRUCTURE.

IT:Network:Applications VIRTUAL DESKTOP INFRASTRUCTURE.
Alcatel Customized Training Site David Otero University of San Diego MSIT 526 Dr. Carl Rebman.

Alcatel Customized Training Site David Otero University of San Diego MSIT 526 Dr. Carl Rebman.
VIRTUALISATION OF HADOOP CLUSTERS Dr G Sudha Sadasivam Assistant Professor Department of CSE PSGCT.

VIRTUALISATION OF HADOOP CLUSTERS Dr G Sudha Sadasivam Assistant Professor Department of CSE PSGCT.
A Network-aware Job Management System Leveraging Software Defined Networking Yasuhiro Watashiba Cybermedia Center, Osaka University, Japan 2014/04/11PRAGMA261.

A Network-aware Job Management System Leveraging Software Defined Networking Yasuhiro Watashiba Cybermedia Center, Osaka University, Japan 2014/04/11PRAGMA261.
Assoc. prof., dr. Vladimir Dimitrov University of Sofia, Bulgaria

Assoc. prof., dr. Vladimir Dimitrov University of Sofia, Bulgaria
System Center 2012 Setup The components of system center App Controller Data Protection Manager Operations Manager Orchestrator Service.

System Center 2012 Setup The components of system center App Controller Data Protection Manager Operations Manager Orchestrator Service.
Developing Reusable Software Infrastructure – Middleware – for Multiscale Modeling Wilfred W. Li, Ph.D. National Biomedical Computation Resource Center.

Developing Reusable Software Infrastructure – Middleware – for Multiscale Modeling Wilfred W. Li, Ph.D. National Biomedical Computation Resource Center.
 Cloud computing  Workflow  Workflow lifecycle  Workflow design  Workflow tools : xcp, eucalyptus, open nebula.

 Cloud computing  Workflow  Workflow lifecycle  Workflow design  Workflow tools : xcp, eucalyptus, open nebula.
CS492: Special Topics on Distributed Algorithms and Systems Fall 2008 Lab 3: Final Term Project.

CS492: Special Topics on Distributed Algorithms and Systems Fall 2008 Lab 3: Final Term Project.

Similar presentations

Ads by Google