Seed Working Group Swiss National Grid Association (SwiNG) [email protected]
-
Upload
caleb-holmes -
Category
Documents
-
view
28 -
download
1
description
Transcript of Seed Working Group Swiss National Grid Association (SwiNG) [email protected]
May 20, 2008
1CCGrid 2008, Lyon, France
Initializing a National Grid Infrastructure: Lessons Learned from the Swiss National
Grid Association Seed Project
Seed Working GroupSwiss National Grid Association (SwiNG)
May 20, 2008
2CCGrid 2008, Lyon, France
Members of the Seed Working Group
Nabil Abdennadher, Haute École Spécialisée de Suisse Occidentale (HES-SO)
Peter Engel, University of Bern (UniBE)
Derek Feichtinger, Paul Scherrer Institute (PSI)
Dean Flanders, Friedrich Miescher Institute (FMI)
Placi Flury, SWITCH
Sigve Haug, University of Bern (UniBE)
Pascal Jermini, École Polytechnique Fédérale de Lausanne (EPFL)
Sergio Maffioletti, Swiss National Supercomputing Centre (CSCS)
Cesare Pautasso, University of Lugano (USI)
Heinz Stockinger, Swiss Institute of Bioinformatics (SIB)
Wibke Sudholt, University of Zurich (UZH) – Chair
Michela Thiemard, École Polytechnique Fédérale de Lausanne (EPFL)
Nadya Williams, University of Zurich (UZH)
Christoph Witzig, SWITCH
May 20, 2008
3CCGrid 2008, Lyon, France
Outline
Background• Grid projects
• SwiNG
Seed Project• Introduction
• Middleware
• Applications
Conclusions and outlook
May 20, 2008
4CCGrid 2008, Lyon, France
Grid Projects and InfrastructuresInternational Grid projects• EGEE (Enabling Grids for E-sciencE): 91 partners,
PRAGMA (Pacific Rim Applications and Grid Middleware Assembly): 29 partners,etc.
National Grid projects• Open Science Grid (USA), ChinaGrid, NAREGI (Japan), e-Science Programme (UK), D-
Grid (Germany), Austrian Grid, etc.
Domain-specific Grid projects• LCG, Chemomentum, GRIDCHEM, Swiss Bio Grid, EMBRACE, DEGREE, etc.
Local Grid projects• XtremWeb-CH, JOpera, etc.
Homogeneous Grid middleware• gLite, UNICORE, Globus, ARC, etc.
May 20, 2008
5CCGrid 2008, Lyon, France
Situation in Europe
Funding for Grid projects by the EU• Within FP5 / FP6 / FP7
• Collaboration projects
National Grid Initiatives (NGIs)• In most European countries
• Some with considerable funding
European Grid Initiative (EGI)• http://web.eu-egi.eu/
• Design study under way
• Following the model of the National Research Networks (NRENs)
May 20, 2008
6CCGrid 2008, Lyon, France
National Grid Initiative (NGI)
Must • Have a mandate to represent
researchers and institutions in Grid-related matters towards
— International bodies (e.g., EU)— Funding agencies— Federal government (SBF, BBT)
• Have only one NGI per country
May• Involve only coordination
• Develop and operate national Grid infrastructure(s)
• Be a legal entity on its own
• Be limited to academic or research institutions
• Also involve participation by the industry
“Coordinating body” for Grid activities within a nation
May 20, 2008
7CCGrid 2008, Lyon, France
Grid in Switzerland before SwiNG
Various, somewhat isolated efforts in the Swiss higher education sector • Some projects within individual
research groups
• Some projects between a limited number of Swiss partners
• Participation in EU-sponsored projects by some institutions
• Participation in international projects by some institutions
No national coordinationNo dedicated fundingNo homogeneous Grid
middleware or infrastructure
May 20, 2008
8CCGrid 2008, Lyon, France
Swiss National Grid Association (SwiNG)
Mission• Ensure competitiveness of Swiss science, education and industry by creating
value through resource sharing.• Establish and coordinate a sustainable Swiss Grid infrastructure, which is a
dynamic network of resources across different locations and administrative domains.
• Provide a platform for interdisciplinary collaboration to leverage the Swiss Grid activities, supporting end-users, researchers, industry, education centres, resource providers.
• Represent the interests of the national Grid community towards other national and international bodies.
History• Initialized in September 2006• Founded as association in May 2007• Operational since January 2008
May 20, 2008
9CCGrid 2008, Lyon, France
Organisational Structure
May 20, 2008
10CCGrid 2008, Lyon, France
Institutional MembersETH domain• École Polytechnique Fédérale de
Lausanne (EPFL)• Eidgenössische Technische Hochschule
Zürich (ETHZ)• ETH Research Institutions (EAWAG,
EMPA, PSI, WSL)• Swiss National Supercomputing Centre
(CSCS)
Cantonal universities• Universität Basel (UniBas)• Universität Bern (UniBE)• Université de Geneve (UniGE)• Université de Neuchâtel (UniNE)• Université de Lausanne (UNIL)• Università della Svizzera Italiana (USI)• Universität Zürich (UZH)
Universities of Applied Sciences• Berner Fachhochschule (BFH)
• Fachhochschule Nordwestschweiz (FHNW)
• Haute Ecole Spécialisée de Suisse Occidentale (HES-SO)
• Hochschule Luzern (HSLU)
• Scuola Universitaria Professionale della Svizzera Italiana (SUPSI)
Specialized institutions• Friedrich Miescher Institute (FMI)
• Swiss Institute of Bioinformatics (SIB)
• Swiss Academic and Research Network (SWITCH)
May 20, 2008
11CCGrid 2008, Lyon, France
Working Groups
Initial WGs• Mandate Letter
• Seed Project— Founded in November 2006— Finished in November 2007
Currently active WGs• ATLAS: High energy physics
• Proteomics: Bioinformatics
• Infrastructure & Basic Grid Services— Grid Architecture Team (GAT)— Grid Operations Team (GOT)— Data Management Team (DMT)
• Education & Training
WGs in planning• AAA/SWITCH projects
• Grid Workflows
• Industry Relations
May 20, 2008
12CCGrid 2008, Lyon, France
Seed Project Working Group Goals
1. Identify which resources (people, hardware, middleware, applications, ideas) are readily available and represent strong interest among the current SwiNG partners.
2. Based on available resources, propose one or more Seed Projects that will help to initialize, test, and demonstrate the SwiNG collaboration. The Seed Project should be realizable in a fast, easy and inexpensive manner (“low hanging fruit”).
3. Help with the coordination and realization of the defined Seed Project.
May 20, 2008
13CCGrid 2008, Lyon, France
Seed Project Survey
Informal inventory of resources available for the Seed Project• Member groups
• Available personnel
• Computer hardware
• Lower-level grid middleware
• Higher-level grid middleware
• Scientific application software and data
• Seed project ideas
12 answers in December 2006
Results• There is a lot of interest and expertise.
• There is enough hardware available, but no direct funding for people.
• Middleware and applications are diverse, but some are more common.
• Main interest is in specific tools and Grid interoperability.
Build a cross-product/matrix infrastructure of selected Grid middleware and applications by gridifying each application on each middleware pool in a non-intrusive manner
Avoid “chicken-and-egg” dilemma in bootstrapping a Grid infrastructure by using known tools and addressing early adopters
May 20, 2008
14CCGrid 2008, Lyon, France
Selection Process
Middleware• Criteria
— Already deployed at partner sites— Sufficient expertise and manpower— Supported within existing larger Grid
efforts— Not too complex requirements— Must be diverse and provide sufficient
set of capabilities
• Initial focus— EGEE gLite (deployed at CSCS, PSI,
SIB, SWITCH, UniBas)— Nordugrid ARC (deployed at CSCS,
SIB, UniBas, UniBE, UZH)— XtremWeb-CH (developed and deployed
at HES-SO)— Condor (deployed at EPFL)
Applications• Criteria
— Need from the Swiss scientific user community
— Computational demand warrants Grid execution
— Sufficient expertise and manpower— Not too complex requirements— Simple gridification, without changing
the source code if possible— Should be diverse and cover sufficient
set of requirements— Reusage of existing Grid-enabled
applications
• Initial focus— Cones (mathematical crystallography,
individual code)— GAMESS (quantum chemistry, standard
free open source code) Huygens (remote deconvolution for
imaging, standard commercial code)— PHYLIP (bioinformatics, standard free
open source code)
May 20, 2008
15CCGrid 2008, Lyon, France
Seed Project Definition
First focus
gLite(Globus
2)CondorARC
XtremWeb-CH
UnitedDevices
Globus4
(WSRF)
Lower-level middleware systems
Meta-middleware and grid interoperability
First focus
Stan-dards
SecurityImple-menta-
tion
Require-ments
Manage-ment
Testing
First focus
Scientific application software
GAMESS
PHYLIP MascotConesHuy-gens
…Swiss
BioGridPhysics
Produc-tion
…
UNICORE
…
May 20, 2008
16CCGrid 2008, Lyon, France
Grid Security: SWITCHslcsSWITCH Short Lived Credential Service (SLCS)• http://www.switch.ch/grid/slcs/
• Ad-hoc generated X.509 certificates
• Based on SWITCHaai (Authentication and Authorization Infrastructure)
• EUGridPMA accredited
• Valid 1’000’000 seconds (ca. 11 days)
• Java-based client software
Advantages• The user does not have to keep track of
where he/she copied his/her certificates between hosts.
• He/she only needs to use his/her SWITCHaai federation account to obtain a certificate, thus he/she has to maintain one credential less.
• He/she does not have to take care of the expiration, respectively, renewal of the certificates. He/she simply requests a new one.
• Identity management becomes simpler since the central Certification Authority (CA) is not required to keep a separate master user database.
• As the SLCS is accredited by the International Grid Trust Federation (IGTF), the certificate is recognized by all Grid resources where the IGTF certificate bundle is installed.
Achievements in Seed Project• Testing on gLite, ARC and Condor pools• MyProxy server for automatic renewal of
expiring proxy certificates to bridge long-running jobs
May 20, 2008
17CCGrid 2008, Lyon, France
Middleware: EGEE gLiteDeployment status in Switzerland• In production in LCG and DILIGENT projects since
2002
• SWITCH (Zurich): Resource Broker, VOMS, CE/UI locally behind firewall
• CSCS (Manno): CE, SE, UI
• SIB (Lausanne): UI
• UZH (Zurich): UI
Achievements in Seed Project• Small test-bed with all essential services, but no big
computer resources• Creation of new Virtual Organization (VO)• Working with SWITCHslcs
• Testing with Cones application
• Access to EGEE resources possible
Disadvantages• Complex due to rich functionality, many vendors, and
partly competing implementations
• Installing and running the UI is straightforward, but large efforts and manpower required for installing and running service components
• Comparatively intrusive on resources (e.g., requires Scientific Linux, workers on compute nodes)
Middleware of the world’s largest Grid infrastructure• http://www.glite.org/• Grid middleware developed and deployed in the EGEE
project, installed in most European countries, and used for CERN’s LCG project
• Based on Globus, adding VO support and extending data management functionalities
• Computing Elements (CEs) interfacing worker nodes to LRMS, Storage Elements (SEs) providing standardized data access and transfer services, information system and resource management, User Interface (UI) client
• Offers many Data Grid components (e.g., file catalogues, storage management)
• Security based on GSI with VOMS (VO Membership Service) support, including hierarchical subgroups
• Several different flavours, running on Scientific Linux
May 20, 2008
18CCGrid 2008, Lyon, France
Middleware: NorduGrid ARCAdvanced Resource Connector• http://www.nordugrid.org/middleware/• Grid middleware developed and deployed in the
NorduGrid project of the Nordic countries• Enables production-quality grids, including information
services, resource, job and data management• Uses replacements and extensions of Globus pre-WS
services (e.g., GridFTP)• Cluster-of-clusters model, Computing and Storage
Elements (CEs and SEs), application Runtime Environments (REs)
• Security based on GSI with VOMS support• Open source under GPL license, supports up to 22
different Unix distributions
Deployment status in Switzerland• Originally deployed as part of the LHC and Swiss Bio
Grid projects• CSCS (Manno): GIIS, CE, SE• UZH (Zurich): CE (10 node cluster)• EPFL (Lausanne): CE (Condor pool)• SIB (Lausanne): CE
Achievements in Seed Project• Configuration of resources at CSCS and UZH,
interfacing of Condor pool at EPFL• Working with SWITCHslcs• Deployment and testing of Cones and GAMESS
applications, running of Cones in production by scientific user
Most successful middleware pool, non-intrusive solution
Disadvantages• Only limited support for complex data management
(e.g., no notion of data proximity)• Compute nodes usually expected to have shared file
system• Information service limited and not very scalable• Coordination among sites necessary for stable
configuration, application REs, and error tracking
May 20, 2008
19CCGrid 2008, Lyon, France
Middleware: XtremWeb-CHHigh-performance Desktop Grid / volunteer computing / P2P middleware• http://www.xtremwebch.net/
• Developed by Nabil Abdennadher et al. at HES-SO
• For deployment and execution on public, non-dedicated platforms via user participation
• Symmetric model of providers and consumers
• Supports direct communication of jobs between compute nodes, also across firewalls
• Can fix the granularity of the application according to the state of the platform
Functionalities• Four modules: Coordinator, worker, warehouse, and
broker
• Volatility of workers
• Automatic execution of parallel and distributed applications
• Direct communication between workers, pull model
• Load balancing
Deployment status in Switzerland• Ca. 200 workers (mainly Windows, few Linux
platforms)
• Sites: EIG (Geneva), HEIG-VD (Yverdon)
Achievements in Seed Project• Test installation at UZH
• PHYLIP application deployed previously
• Integration of GAMESS application
Disadvantages• Security limited and based on central user database,
not compatible with GSI, VOs, and SWITCHslcs
• Porting of applications needs some effort
• No special data management features
May 20, 2008
20CCGrid 2008, Lyon, France
Middleware: CondorHigh-throughput computing environment• http://www.cs.wisc.edu/condor/
• Provides infrastructure for volatile Desktop Grid resources, also cross-institutional
• Several authentication and authorisation mechanisms (e.g., GSI, Kerberos)
• Job queue and resource management for fair and optimized assignment and sharing
• Shared file system or input/output file transfer to/from the compute nodes
• Multi-platform (Linux, Windows, Mac OS X, some other Unix variants), open source
• Can be interfaced with other middleware (e.g., UNICORE, Globus, ARC) as LRMS
Deployment status in Switzerland• Existing production pool at EPFL:
http://greedy.epfl.ch/
• Ca. 200 desktop CPUs (60% Windows, 40% Linux or Mac OS X machines), behind firewall
• Computing power available only during nights and weekends, machine owner has priority
• One submit server and one central manager
• No access to compute nodes for third-party software installation (Condor installed by node owners, not Grid managers), thus built-in file transfer protocol required to transport application binaries along with input data
• Due to desktop nature relatively short jobs advised (6 h max., not enforced)
Achievements in Seed Project• Interfacing to ARC pool
• Working with SWITCHslcs
• Testing of Cones and GAMESS applications
May 20, 2008
21CCGrid 2008, Lyon, France
Middleware Interoperability
Despite existing OGF standards such as JSDL (Job Submission Description Language), most middleware systems have their own mechanisms for resource and data management, information representation, or job submission.
Solutions for interoperability1. Meta-middleware: Complex due to different interfaces and missing standards,
thus out of scope for the Seed Project
2. One-to-one wrappers: Some middleware as entry point and bridge, transforming one format to another
Achievements in Seed Project• Integration of Condor pool in ARC pool based on existing wrapper
• ARC installed on gateway machine
• Modified to allow transparent appending of required binaries
May 20, 2008
22CCGrid 2008, Lyon, France
Seed Project Definition
First focus
gLite(Globus
2)CondorARC
XtremWeb-CH
UnitedDevices
Globus4
(WSRF)
Lower-level middleware systems
Meta-middleware and grid interoperability
First focus
Stan-dards
SecurityImple-menta-
tion
Require-ments
Manage-ment
Testing
First focus
Scientific application software
GAMESS
PHYLIP MascotConesHuy-gens
…Swiss
BioGridPhysics
Produc-tion
…
UNICORE
…
May 20, 2008
23CCGrid 2008, Lyon, France
Application: Cones
Mathematical crystallography program• For given representative quadratic form,
calculates its subcone of equivalent combinatorial types of parallelohedra
• For dimension d = 6, number expected to be greater than 200’000’000 (currently 161’299’100)
Code properties• Single-threaded C program• Developed by Peter Engel, UniBE• Several text input files, one execution
command, several text output files
Possibilities for Grid distribution• Running of several jobs off the same
input file• Cutting of input file into pieces
Achievements in Seed Project• Refactoring of source code• Creation of configure and make files• Testing on gLite, ARC and Condor
pools• Running in production on ARC pool
with first scientific user• Ca. 50’000 new combinatorial types of
primitive parallelohedra identified• Still a lot of room for improvement
regarding ease and efficiency of use
May 20, 2008
24CCGrid 2008, Lyon, France
Application: GAMESSGeneral Atomic and Molecular Electronic Structure System• http://www.msg.chem.iastate.edu/gamess/
• Program package for ab initio molecular quantum chemistry
• Computing of molecular systems and reactions in gas phase and solution (properties, energies, structures, spectra, etc.)
• Wide range of methods for approximate solutions of the Schrödinger equation from quantum mechanics
• Standard free open source code developed and used by many groups (e.g., at UZH)
Code properties• Mainly Fortran 77 and C code and shell scripts• Available for large variety of hardware architectures
and operating systems• Usually one keyword-driven text input file, one
execution command, several text output files• Well parallelized by its own implementation, called
Distributed Data Interface (DDI)• Comes with more than 40 functional test cases
Possibilities for Grid distribution• External: Embarrassingly parallel parameter scans in
input file• Internal: Component distribution based on current DDI
parallelization implementation
Achievements in Seed Project• Deployment and testing on ARC and Condor pools at
CSCS, EPFL, and UZH• Integration into XtremWeb-CH• Simple corannulene DFT
functional scan test caseprovided by Laura Zoppi,UZH
• 223 small molecule MP2calculations test case provided by Kim Baldridge,UZH
May 20, 2008
25CCGrid 2008, Lyon, France
Application: PHYLIPPHYLogeny Inference Package• http://evolution.genetics.washington.edu/phylip
.html
• Used to generate “life trees” (evolutionary trees, interfering phylogenies)
• Most widely distributed phylogeny package, in development since the 1980s, 15’000 users
• Tree composed of several branches, subbranches, and leaves (sequences), which are complex and CPU-intensive to construct, compare, and select
“Life tree”
Code properties• Package of ca. 34 program modules• C source code and executables (Windows,
Mac OS X, Linux) available• Input data read from text files, data
processed, output data written onto text files
• Data types: DNA sequences, protein sequences, etc.
• Methods available: Parsimony, distance matrix, likelihood methods, bootstrapping, and consensus trees
Possibilities for Grid distribution• Workflows constructed by users• Distribution following Single Program
Multiple Data (SPMD) modelAchievements in Seed Project
• Integration and deployment on XtremWeb-CH as independent project (Seqboot, Dnadist, Fitch-Margoliash, Neighbor-Joining and Consensus modules)
• Parallel version of Fitch module• Execution of HIV sequences-related test
case on XtremWeb-CH pool• Web service for dynamic configuration of
application platform and parameters
May 20, 2008
26CCGrid 2008, Lyon, France
Overview of AchievementsEGEE gLitemiddleware
NorduGrid ARC middleware
XtremWeb-CH middleware
Condor middleware
Pool established CSCS, SIB, and SWITCH, VO created, UI at UZH
CSCS, SIB, and UZH, from Swiss Bio Grid
HES-SO and UZH EPFL, coupling with ARC
SLCS security Tested at CSCS, SWITCH, and SIB
Tested at UniBE and UZH
Needs changes in or interface to middleware
Tested at EPFL, via ARC
Conesapplication
Tested at SIB Scientific usage from UniBE
Not started yet Tested at EPFL
GAMESS application
Not started yet Test usage from UZH
Work in progress at HES-SO
Tested at EPFL
Huygens application
No personnel or license
No personnel or license
No personnel or license
No personnel or license
PHYLIP application
Not started yet Not started yet Preexisting at HES-SO
Work in progress at EPFL
May 20, 2008
27CCGrid 2008, Lyon, France
Lessons LearnedProject• Approach based on heterogeneous set of Grid middleware and applications turned out to be useful to
initialize technical collaboration.• There is considerable interest and expertise in Grid resource and knowledge sharing in Switzerland,
which previously has been directed mostly towards external projects.• Dedicated partners, clear responsibilities, continuous communication, detailed documentation, and
active project management are required.• Funding, in particular for people, has to be properly secured for production setup.
Middleware• Middleware is still demanding to install, maintain, and use, mainly due to its complexity and insufficient
documentation.• Middleware architectures and interfaces differ considerably, and require efforts in interoperability.• There is strong need for simplification and standardization, both technically and regarding procedures
(e.g., for policy, development, deployment, and execution mechanisms).
Applications• Applications are diverse and can be put onto the Grid in different ways, and therefore need direct
cooperation among scientific developers and Grid experts, that is, interdisciplinary work.• Scientists are mainly interested in the implementation of new methods, thus standards for software
development and packaging are often ignored, leading to poor installation, documentation, and sometimes performance.
• To run applications on the Grid, applications need to provide well documented and packaged distributions including standardized installation, configuration, testing and use procedures.
May 20, 2008
28CCGrid 2008, Lyon, France
OutlookFuture topics
• Setup of a production infrastructure• Inclusion of additional resources• Extension to new applications• Evaluation of further middleware• Expansion to data management• Connection to other Grid infrastructures
Continuation of work in Switzerland• SwiNG Working Groups• SwiNG-related, funded projects• Securing of dedicated funding for SwiNG
General focus• Standardisation and interoperation of middleware• Professionalisation and standardisation of applications
May 20, 2008
29CCGrid 2008, Lyon, France
Thank you!
Questions?