MPI-VT within EGI Costantini Alessandro INFN — IGI

MPI-VT within EGI Costantini Alessandro INFN - IGI alessandro. costantini@cnaf. infn. it 23/03/2012 Costantini A. – MPI-Multicore 1

Virtual Team Activity established within the EGI-In. SPIRE project - user engagement activity - started in 2011 Based on two key elements Virtual Team framework NGI International Liaisons 23/03/2012 Costantini A. – MPI-Multicore 2

Virtual Team framework: The Virtual Team framework enables NGI and EGI. eu persons to initiate and participate in short living projects (Virtual Team projects) that focus on well defined, non-operational activities around the production infrastructure. 23/03/2012 Costantini A. – MPI-Multicore 3

Virtual Team NGI International Liaisons: The interaction between the NGI teams and EGI. eu on nonoperational activities is undertaken through an "NGI International Liaisons"(NILs in short), who are responsible within the NGIs for the delivery and interaction of non-operational tasks. 23/03/2012 Costantini A. – MPI-Multicore 4

MPI-VT: General informations - Leader: Alvaro Simon (CESGA, Spain) Zdenek Sustr (CESNET, Czech Republic) - Start Date: 10/Nov/2011 - End Date: 31/May/2012 23/03/2012 Costantini A. – MPI-Multicore 5

Motivations Despite a dedicated SA 3 activity to support MPI there still seem to be significant issues in uptake and satisfaction amongst the user communities. This VT - Works with user communities and projects that use MPI resources - Improve the communication between MPI users and developers of MPI support within EGI SA 3. 23/03/2012 Costantini A. – MPI-Multicore 6

Members CZ: Zdenek Sustr (leader) ES/IBERGRID: Alvaro Simon (leader), Enol Fernandez, Iván Díaz, Alvaro Lopez, Pablo Orviz, Isabel Campos, Roberto Rosende Dopazo GR: Dimitris Dellis, Marios Chatziangelou, Paschalis Korosoglou HR: Emir Imamagic, Luko Gjenero IE: John Walsh IT: Daniele Cesini, Alessandro Costantini, Vania Boccia, Marco Bencivenni PT: Gonçalo Borges SK: Viera Sipkova, Viet Tran, Jan Astalos UK: John Gordon EGI. eu: Gergely Sipos, Karolis Eigelis, Tiziana Ferrari, Peter Solagna 23/03/2012 Costantini A. – MPI-Multicore 7

Tasks Task 1: MPI documentation Task 2: Nagios probes Task 3: Information system Task 4: Accounting system Task 5: Batch system status Task 6: Gather information from MPI sites 23/03/2012 Costantini A. – MPI-Multicore 8

Task 1: MPI documentation Review EGI documentation - Unser guide - Admin Guide 23/03/2012 Costantini A. – MPI-Multicore 9

Task 2: Nagios probes Review the current nagios probes MPI nagios new specifications: - org. sam. mpi. Env. Sanity. Check -- Test the information published by the (MPI or Parallel) service Glue. CEPolicy. Max. Slots. Per. Job - org. sam. mpi. Simple. Job -- Test the MPI functionality with a minimum set of resources (Cpu. Number = 2; Node. Number=2) - org. sam. mpi. Complex. Job -- Test the MPI functionality (Cpu. Number = 4; Node. Number=2; SMPGranularity=2; Whole. Nodes=True) 23/03/2012 Costantini A. – MPI-Multicore 10

Task 3: Information system - detecting MPI resources Max. Slots. Per. Jobs (GLUE 2) value is not filled by the current LRMS Information Providers cput = Maximum amount of CPU time used by all processes in the job pcput = Maximum amount of CPU time used by any single process in the job walltime = Maximum amount of real time during which the job can be in the running state. 23/03/2012 Costantini A. – MPI-Multicore 11

Task 4: Accounting system - Implement MPI accounting system No special accounting up to date - Contact with APEL and Accounting Portal 23/03/2012 Costantini A. – MPI-Multicore 12

Task 5: Batch system status - Two bugs were found during the first UMD verification of WN/Torque + EMI-MPI. 1. 0. Is a torque/maui problem that affects all MPI jobs. Maui versions prior to 3. 3. 4 do not allocate correctly all the nodes for the job execution - IGI has presented a possible workaround to the problem for Maui versions prior to 3. 3. 4 - Bug fixed in EMI 1 -IGI (full fixed in EMI 2) 23/03/2012 Costantini A. – MPI-Multicore 13

Task 6: Gather information from MPI sites - MPI VO -- bring together sites and users interested in MPI - This VO is NOT intended for everyday use by all users wishing to use MPI - This VO IS intended for users who wish to cooperate with the VT to make MPI support in EGI better - The main reason for its establishment is to collect experience that will be later adopted by regular VOs 23/03/2012 Costantini A. – MPI-Multicore 14