GridLab Resource Management System (GRMS) Jarek Nabrzyski GridLab Project Coordinator Poznań Supercomputing and Networking Center
GlobusWorld, San Diego 2003 GridLab EU funded project, involving 11 European and 3 American partners (Globus and Condor teams), January 2002 – December 2004 Main goal: to develop a Grid Application Toolkit (GAT) and set of grid services and tools... resource management (GRMS), data management, monitoring, adaptive components, mobile user support, security services, portals,... and test them on a real testbed with real applications
GlobusWorld, San Diego 2003 What our users want... Two primary applications: Cactus and Triana other application communities are also being engaged, Application oriented environment (whatever it is) Resources (grid) on demand Adaptive applications – adaptive grid environment job checkpoint, migration, spawn off a new job when needed, Open, pervasive, not even restricted to a single Virtual Organization The ability to work in a disconnected environment start my job on a disconnected laptop; migrate it to grid when it becomes available from laptops to fully deployed Virtual Organisations Mobile working Security
GlobusWorld, San Diego 2003 What our users want... (cont.) The infrastructure must provide capabilities to customise choice of service implementation (e.g. using efficiency, reliability, first succeeding, all) Advance reservation of resources To be able to express their preferences regarding their jobs on one hand and to understand the resources’ policies on the other Policy information and negotiation mechanisms what is a policy of usage of your resources? Prediction-based information How long will my job run on particular resource? What resources do I need to complete the job before deadline?
GlobusWorld, San Diego 2003 GridLab RMS approach Grid resources are not only the machines, but also databases, files, users, administrators, instruments, mobile devices, jobs/applications... Many metrics: throughput, cost, latency, deadline, other time and cost ones... Grid resource management consists of job/resource scheduling, security (authorization services,...), local policies, negotiations, accounting,... GRM is both, user and resource owner driven negotiation process and thus, multicriteria decision making process
GlobusWorld, San Diego 2003 GridLab RMS Job Receiver Jobs Queue BROKER Execution Unit Resource Discovery Scheduler Resource Reservation Prediction Unit File Transfer Unit Information Services Data Management Authorization System Adaptive Workflow Manager SLA Negotiation GRMS Monitoring GLOBUS, other Local Resources (Managers)
GlobusWorld, San Diego 2003 GRMS and SLA
GlobusWorld, San Diego 2003 GRMS and SLA (cont.)
GlobusWorld, San Diego 2003 Does GRAM-2 solve our GRM goals? SLA is a „has to be” for our users GRAM-2 vs. local policies/resource management systems Is a mapping of SNAP onto the local schedulers straightforward? Is a technology there in the local schedulers to support SNAP? GRAM-2 solves many of our problems, but still additional grid services need to be developed and new features of GRAM would be welcome
GlobusWorld, San Diego 2003 Needed work/research Focus on the infrastructure is not enough for the efficient GRM Focus on policies Focus on multicriteria aspects of the GRM users, their preferences and applications resource owners’ preferences preference models, multicriteria decision making, knowledge will be crucial for efficient resource management Focus on AI techniques for GRM (and not only) Focus on business models, economy grids Cost negotiation mechanisms could be part of the SLA negotiation process contradictory in nature