SEEGRID Nis Training Session Introduction to Grids SEEGRID

SEE-GRID Nis Training Session Introduction to Grids SEE-GRID Basic Facts Aleksandar Belić SCL, Institute of Physics, Belgrade Jan 31, 2006

Science and technology are team sports AEGIS Jan 31, 2006 SEE-GRID Nis Training Session

Unifying concept: Grid AEGIS Resource sharing and coordinated problem solving in dynamic, multi-institutional virtual organizations. Jan 31, 2006 SEE-GRID Nis Training Session

What types of problems is the Grid intended to address? AEGIS Jan 31, 2006 n Too hard to keep track of authentication data (ID/password) across institutions n Too hard to monitor system and application status across institutions n Too many ways to submit jobs n Too many ways to store & access files/data n Too many ways to keep track of data n Too easy to leave “dangling” resources lying around (robustness) SEE-GRID Nis Training Session

Requirements n n AEGIS n n Security Monitoring/Discovery Computing/Processing Power Moving and Managing Data Managing Systems System Packaging/Distribution What end users need? Secure, reliable, on-demand access to data, software, people, and other resources (ideally all via a Web Browser!) Jan 31, 2006 SEE-GRID Nis Training Session

Ingredients for GRID development AEGIS n n Right balance of push and pull factors is needed Supply side n n Demand side n n Jan 31, 2006 Technology – inexpensive HPC resources (linux clusters) Technology – network infrastructure Financing – domestic, regional, EU, donations from industry Need for novel e. Science applications Hunger for number crunching power and storage capacity SEE-GRID Nis Training Session

Supply side - clusters n AEGIS n The cheapest supercomputers – massively parallel PC clusters This is possible due to: n n n Advantages: n n n n Good knowledge of parallel programming is required Hardware needs to be adjusted to the specific application (network topology) More complex administration Tradeoff: brain power purchasing power The next step is GRID: n n Jan 31, 2006 Widespread choice of components/vendors, low price (by factor ~5 -10) Long warranty periods, easy servicing Simple upgrade path Disadvantages: n n Increase in PC processor speed (> Gflop/s) Increase in networking performance (1 Gbs) Availability of stable OS (e. g. Linux) Availability of standard parallel libraries (e. g. MPI) Distributed computing, computing on demand Should “do for computing the same as the Internet did for information” (UK PM, 2002) SEE-GRID Nis Training Session

Country-wide inventory of Clusters (Example: Serbia) n AEGIS n n n n Jan 31, 2006 I. of Phys. – SCL, 64+2 x P 4 Xeon 2. 8 GHz Hydro-Meteo. I. , 20+1 x Atlon 1. 4 GHz F. Electronic Eng. Niš, 8 x P 4 1. 8 GHz F. Agriculture NS, 16 x P 4 2. 4 GHz SEWA Beograd, 8 x P 4 2. 2 GHz Center for supercomp. Kg, 16 x P 4 1. 8 GHz F. Mechanical Eng. Bgd, 8 x P 3 1 GHz F. Electrical Eng. Bgd, 8 x P 3 1 GHz + 16 x P 4 Xeon 2. 8 GHz Several scavenger sites SEE-GRID Nis Training Session

Supply side - network n AEGIS Needed at all scales: n n n n Jan 31, 2006 World-wide Pan-European (GEANT 2) Regional (SEEREN 2, …) National (NREN) Campus-wide (WAN) Building-wide (LAN) Remember – it is end user to end user connection that matters SEE-GRID Nis Training Session

GÉANT 2 Pan-European IP R&E network AEGIS Jan 31, 2006 SEE-GRID Nis Training Session

GÉANT 2 Global Connectivity AEGIS Jan 31, 2006 SEE-GRID Nis Training Session

Future development of regional network AEGIS Jan 31, 2006 SEE-GRID Nis Training Session

NREN (example: Serbia) AEGIS Jan 31, 2006 SEE-GRID Nis Training Session

Supply side - financing n AEGIS National funding (Ministries responsible for research) n n n Bilateral projects and donations Regional initiatives n n n Jan 31, 2006 Networking (HIPERB) Action Plan for R&D in SEE EU funding n n Lobby gvnmt. to commit to Lisbon targets Level of financing should be following an increasing trend (as a % of GDP) Seek financing for clusters and network costs FP 6 – IST priority, e. Infrastructures & GRIDs CARDS Other international sources (NATO, …) Donations from industry (HP, SUN, …) SEE-GRID Nis Training Session

FP 6 AEGIS Jan 31, 2006 SEE-GRID Nis Training Session

Demand side - e. Science n Usage of computers in science: n AEGIS n n n Why is the use of computation in science growing? n n Jan 31, 2006 Trivial: text editing, elementary visualization, elementary quadrature, special functions, . . . Nontrivial: differential eq. , large linear systems, searching combinatorial spaces, symbolic algebraic manipulations, statistical data analysis, visualization, . . . Advanced: stochastic simulations, risk assessment in complex systems, dynamics of the systems with many degrees of freedom, PDE solving, calculation of partition functions/functional integrals, . . . Computational resources are more and more powerful and available (Moore’s law) Standard approaches are having problems Experiments are more costly, theory more difficult Emergence of new fields/consumers – finance, economy, biology, sociology Emergence of new problems with unprecedented storage and/or processor requirements SEE-GRID Nis Training Session

Demand side consumers n Those who study: n AEGIS n n n n n Who can deliver? Those with: n n n Jan 31, 2006 Complex discrete time phenomena Nontrivial combinatorial spaces Classical many-body systems Stress/strain analysis, crack propagation Schrodinger eq; diffusion eq. Navier-Stokes eq. and its derivates functional integrals Decision making processes w. incomplete information … Adequate training in mathematics/informatics Stamina needed for complex problems solving Answer: rocket scientists (natural sciences and engineering) SEE-GRID Nis Training Session

Overview of SEE-GRID & SEE-GRID 2 projects n After a brief pause! AEGIS Jan 31, 2006 SEE-GRID Nis Training Session