REsources linkage for E-scIence - RENKEI -

Similar documents
Application Hosting Services for Research Community on Multiple Grid Environments

NAREGI PSE with ACS. S.Kawata 1, H.Usami 2, M.Yamada 3, Y.Miyahara 3, Y.Hayase 4, S.Hwang 2, K.Miura 2. Utsunomiya University 2

Introduction of NAREGI-PSE implementation of ACS and Replication feature

EGEE and Interoperation

aginfra: High Performance Compu8ng einfrastructure for Agriculture

EOSC Services & Architecture: the EOSC-hub approach Tiziana Ferrari, Project Coordinator, EGI Founda?on

Independent Software Vendors (ISV) Remote Computing Usage Primer

UNICORE Globus: Interoperability of Grid Infrastructures

Presented by Wolfgang Ziegler, Open Grid Forum

FutureGrid 101. Part 2: Ge*ng Started Craig Stewart

NAREGI Middleware Overview V 1.1 March, 2011 National Institute of Informatics

Overview of XSEDE for HPC Users Victor Hazlewood XSEDE Deputy Director of Operations

Workflow applications on EGI with WS-PGRADE. Peter Kacsuk and Zoltan Farkas MTA SZTAKI

Grid Scheduling Architectures with Globus

UAB IT Research Compu3ng Update

Grid Experiment and Job Management

Topics of Discussion

First European Globus Community Forum Meeting

globus online The Galaxy Project and Globus Online

Grid Interoperation and Regional Collaboration

The University of Oxford campus grid, expansion and integrating new partners. Dr. David Wallom Technical Manager

October 5 th 2010 NANOG 50 Jason Zurawski Internet2

How to build Scientific Gateways with Vine Toolkit and Liferay/GridSphere framework

Interoperable job submission and management with GridSAM, JMEA, and UNICORE

CCW Workshop Technical Session on Mobile Cloud Compu<ng

Concepts and Design of an Interoperability Reference Model for Scientific- and Grid Computing Infrastructures

AAI in EGI Current status

Gergely Sipos MTA SZTAKI

Cloud Computing. Up until now

THE WIDE AREA GRID. Architecture

CHAPTER 2 LITERATURE REVIEW AND BACKGROUND

Federated access to e-infrastructures worldwide

Grid Programming: Concepts and Challenges. Michael Rokitka CSE510B 10/2007

Introduc)on to High Performance Compu)ng Advanced Research Computing

Pegasus Workflow Management System. Gideon Juve. USC Informa3on Sciences Ins3tute

INITIATIVE FOR GLOBUS IN EUROPE. Dr. Helmut Heller Leibniz Supercomputing Centre (LRZ) Munich, Germany IGE Project Coordinator

Juliusz Pukacki OGF25 - Grid technologies in e-health Catania, 2-6 March 2009

SZDG, ecom4com technology, EDGeS-EDGI in large P. Kacsuk MTA SZTAKI

QosCosGrid Middleware

Use of Shibboleth by ARCS

Scientific data processing at global scale The LHC Computing Grid. fabio hernandez

Easy Access to Grid Infrastructures

XSEDE Infrastructure as a Service Use Cases

Fujitsu s Activities in Grid Computing

Chapter 2 Introduction to the WS-PGRADE/gUSE Science Gateway Framework

The MEG Metadata Schemas Registry Schemas and Ontologies: building a Semantic Infrastructure for GRIDs and digital libraries Edinburgh, 16 May 2003

ARC NOX AND THE ROADMAP TO THE UNIFIED EUROPEAN MIDDLEWARE

IMPLEMENTING THE WASCAL DATA INFRASTRUCTURE (WADI)

30 Nov Dec Advanced School in High Performance and GRID Computing Concepts and Applications, ICTP, Trieste, Italy

User Community Driven Development in Trust and Identity Services

Real- &me Archiving of Spontaneous Events (Use- Case : Hurricane Sandy)

Computational Web Portals. Tomasz Haupt Mississippi State University

Grid Computing Fall 2005 Lecture 5: Grid Architecture and Globus. Gabrielle Allen

Using MATLAB on the TeraGrid. Nate Woody, CAC John Kotwicki, MathWorks Susan Mehringer, CAC

Strategies to remove complexity from everyday infrastructure

Con$nuous Audi$ng and Risk Management in Cloud Compu$ng

Cybersecurity Curricular Guidelines

South African Science Gateways

Big Data, Big Compute, Big Interac3on Machines for Future Biology. Rick Stevens. Argonne Na3onal Laboratory The University of Chicago

The NAREGI Server Grid Resource Management Framework

The balancing act: Research vs. Produc7on. Kobus Van der Merwe

INSPIRE and Service Level Management Why it matters and how to implement it

NASPInet 2.0 The Evolu4on of Synchrophasor Networks

Data Access and Analysis with Distributed, Federated Data Servers in climateprediction.net

Abstract. 1. Introduction. M. Marzolla. L. Zangrando INFN Padova, Italy

PoS(EGICF12-EMITC2)081

XSEDE Software and Services Table For Service Providers and Campus Bridging

Composite Compliance: Demonstra1ng Suitability of Cloud Layering for Sensi1ve and Regulated Workloads

Parallel computing, data and storage

DataONE Cyberinfrastructure. Ma# Jones Dave Vieglais Bruce Wilson

How the Cloud is Changing Federated Iden4ty Requirements. Patrick Harding CTO, Ping March 1, 2010

Cloud Standards Coordina.on

Jeff Gooding Southern California Edison. Innovation at Southern California Edison

Implementing MPI on Windows: Comparison with Common Approaches on Unix

GROWL Scripts and Web Services

Overview of IPTV Forum Japan s Hybridcast Technical SpecificaAon

Today s Objec4ves. Data Center. Virtualiza4on Cloud Compu4ng Amazon Web Services. What did you think? 10/23/17. Oct 23, 2017 Sprenkle - CSCI325

On-demand Research Computing: the European Grid Infrastructure

Enterprise Architecture CS 4720 Web & Mobile Systems

Distributed Research Networks: Lessons from the Field

MulG-Vendor Key Management with KMIP

EUDAT & AAI. Daan Broeder MPI for Psycholinguistics

Grid Computing. MCSN - N. Tonellotto - Distributed Enabling Platforms

glite Grid Services Overview

Crea?ng Cloud Apps with Oracle Applica?on Builder Cloud Service

Molecular dynamics simulations in the MolDynGrid Virtual Laboratory by means of ARC between Grid and Cloud

The Social Grid. Leveraging the Power of the Web and Focusing on Development Simplicity

IRODS USER GROUP 2014 CAMBRIDGE,MA John Burns. 6/25/14 Archive Analy3cs Solu3ons 1

The Fedlet: Real World Examples

Advanced Research Compu2ng Informa2on Technology Virginia Tech

Concurrency-Optimized I/O For Visualizing HPC Simulations: An Approach Using Dedicated I/O Cores

Cyber Security Capabilities

Grid Computing Middleware. Definitions & functions Middleware components Globus glite

Institute of Cybernetics NAS of Ukraine Valentyna Cherepynets

Integration of Network Services Interface version 2 with the JUNOS Space SDK

UNIT IV PROGRAMMING MODEL. Open source grid middleware packages - Globus Toolkit (GT4) Architecture, Configuration - Usage of Globus

Andrea Sciabà CERN, Switzerland

Eclipse Technology Project: g-eclipse

Interoperation Scenarios of Production e-science Infrastructures

PoS(EGICF12-EMITC2)094

Transcription:

REsources linkage for E-scIence - - hlp://www.e- sciren.org/ REsources linkage for E- science () is a research and development project for new middleware technologies to enable e- science communi?es. "" is Japanese word meaning "federa?on". The goal of the project is to develop middleware to federate/share resources (computers, storages, databases and applica?ons) distributed among mul?ple organiza?ons, such as research laboratories, na?onal computer centers and interna?onal grids. The research ac?vi?es consist of five research themes presented in the figure below. This project is supported by the Ministry of Educa?on, Culture, Sports, Science and Technology in Japan. computa?on intensive applica?ons users database users applica?on developers (1) compu?ng resource federa?on, and applica?on hos?ng (3) federa?on, and federa?on with ID management systems (4) API for mul?ple grid middleware laboratory (LLS) grid middleware (e.g. NAREGI) computer centers (NIS) interna0onal inter- opera0on grid middleware (5) evalua?on and collabora?on with users computer users (2) file sharing, and file catalogue federa?on computa?on/data intensive applica?on users #1143 #1143 #3029 #1143 #3213 #3013 #1143 #1127

-Workflow Tool(WFT) WFT(WorkFlow Tool) provides end users with a single seamless interface to manage workflow jobs. They use WFT to discover appropriate resources, to specify workflow jobs consisting of applications deployed by AHS on various grid resources, to launch jobs, to monitor running jobs, to terminate jobs, to retrieve files to client machine and to transfer files between them with the GUI interface. System features Ø Workflow engine controls over a single workflow execu?on by submi_ng jobs to compu?ng resources on LLS (Laboratory Level System) and NIS (Na?onal Infrastructure System, Super computer centers). Ø Limited communica?on protocols(hlps, myproxy, gfarm) between LLS and NIS resources for communica?ng through firewalls Ø Security Infrastructure (GSI) used as an authen?ca?on mechanism for submi_ng jobs to both resources. User Management Server (UMS) for archiving user cer?ficates and MyProxy Server in NIS are shared between LLS and NIS. Ø File sharing and stage- in/out of files between LLS and NIS by the file system (Gfarm). pre sim 1 sim. 2 sim. 3 post WFT AHS Workflow LLS Portal WF engine Cloud Client SAM WS- GRAM Cloud Nodes LLS Resources Deploy and boot pre post sim 1 SS WF engine Portal (WFT) Super Scheduler NIS Resources(NAREGI) VM VM VOMS,UMS MyProxy sim. 2 sim. 3 IS Gfarm meta server, file server

Interoperation compu?ng enables resource federa?on among organiza?ons. Scien?sts and engineers are ge_ng many benefits from resource sharing among organiza?ons. However, barriers s?ll exist. Because different grid middleware are implemented using different specifica?ons, resources cannot be shared among them. Many grid projects including are making effort to achieve grid interopera?on. User in Another Information Service in another Job Submission Client / Service in Another HPCBP Client HPCBP Service Computing Resource in Another Another Infrastructure -IS Schema Translator -SS SS Services HPCBP-SC HPCBP Client HPCBP-Service HPCBP Server LRMS SC NAREGI VM IS NAS / CDAS NIS Computing Resource Portal NAREGI user -IS LRPS Modification/Addition from NAREGI Infrastructure (NIS) Fig. 1 Interopera?on Architecture Interna?onal standards defined by the OGF (Open Forum) are used for grid interopera?on. " HPCBP: HPC Basic Profile " PGI: Produc?on Infrastructure " BES : Basic Execu?on Service " JSDL: Job Submission Descrip?on Language " GLUE: Informa?on schema etc. Five grid projects in the world tried and succeeded an interopera?on test. Web Server Oxford e-research Centre Pre-Processing: Generate input data (Perl) Main-Processing: Data Staging Job Submission to s Post-Processing-1: Choice Best Result (Perl) Post-Processing-2: Upload results to Web : HPCBP(BES, JSDL, etc.) : FTP or FTP UNICORE (DEISA) SAM (UK-NGS/OMII-UK) ARC (Nordu) Genesis II (U of Virginia) (NII) Local System Environments Internet Fig. 2 A Interopera?on Test The client can submit jobs to resources on different grids in a single workflow. Fig. 3 Using Mul?ple s in a Workflow

Applica0on Hos0ng Service (AHS) AHS provides a repository of applica?on informa?on to the research community created by grid virtual organiza?on (VO). Community members can share the applica?ons and their execu?on environment on hierarchical grid environments as shown in the right Figure. In a research community, applica?on users need to know only the applica?on specifica?ons, such as the applica?on name or keywords for selec?ng the applica?ons. Architecture A Univ. Virtual Organizations Application Developers Groups Application Users Joint Research Community AHS basic architecture is constructed as a set of Web/ service infrastructure on top of WSRF (Web Services Resource Framework) specifica?ons. WSRF is a messaging model and provides the ability to model stateful resources in a framework of Web services. AHS operates coopera?vely with the workflow tool and access to SS (Super Scheduler) for the NIS or SAM for the LLS as shown in the bolom Figure. A Lab. B Lab. B Univ. C Lab. C Univ. DUniv. D Lab. NIS LLS

U0lizing Cloud and for Research and Educa0on NII carries out two projects, edubase Cloud and, to federate cloud and grid infrastructures for research and education. The edubase Cloud offers platforms for educational programs and daily computing resources to educators and researchers. The middleware enable users to run their applications utilizing resources both in a cloud and a grid, e.g. running pre/post es in a cloud (Iaas) and large-scale simulation in a grid. ü plahorms for HPC applica?ons grid cloud service ü plahorms for educa?onal programs service ü plahorms for daily compu?ng ü a gateway to supercompu?ng services educators researchers edubase Cloud prj_educloud.html http://grace-center.jp/en/ http://www.e-sciren.org/index-e.html