1 / 11

Status of Grid-enabled UTA McFarm software

This article provides an update on the status of the Grid-enabled UTA McFarm software, including information on the distribution process, useful documentation, and the future plans for job submission and bookkeeping. It also discusses the status of the components, such as the bookkeeper and McView, a monitoring tool. The article concludes by highlighting the progress made towards building a distributed system for MC production using the Globus toolkit.

emmert
Download Presentation

Status of Grid-enabled UTA McFarm software

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Status of Grid-enabled UTA McFarm software Tomasz Wlodek University of the Great State of TX At Arlington

  2. As a reminder … • In UTA we operate two Linux farms HEP and CSE for MC production • We use McFarm a home-grown batch processing system for D0 MC production • We are currently switching from Linux 6.* to Linux 7.* • In parallel we use the CSE farms to test the scripts for McFarm software installation

  3. A couple of experimental groups in D0 have expressed interest in our software and plan to install it on their farms • LTU, Boston, Tata, Dubna, Brazil, Manchester, Oklahoma, LTU • Of theese Tata, Oklahoma and LTU will become first ones to install McView • We hope that others will follow. We start to distribute McFarm software.

  4. How are we going to distribute the McFarm software? • WWW page http://www-hep.uta.edu/~d0race/McFarm/McFarm.html • You will find there a collection of notes and scripts for installation of farm server, file server, worker nodes, gather servers etc. • Also you will find there additional information: how to install Linux, Globus, Sam, etc. • Software is available for download, but read documentation first!

  5. Useful sites with UTA software documentation: • http://heppc12.uta.edu/~d0race/ : how to install D0 software • http://www-hep.uta.edu/hep_notes/computing.html All UTA related computing documents • http://www-hep.uta.edu/%7Emcfarm/mcfarm/main.htmlMcFarm • http://www-hep.uta.edu/~d0race/McFarm/McFarm.html How to install McFarm

  6. Future of job submission and bookkeeping Only one machine takes care of the job submission and monitoring for all farms! user MC production server www server (production status) Job submission and control via Globus-tools Participating production farms – can be anywhere in the world! SAM

  7. The plan GEM Bookkeeper (Condor-G) McView (MDS) Submitter (Globus,Grid-ftp, Condor-G))

  8. Status of the components: • Bookkeeper exists, can be seen http://heppc1.uta.edu/atlas/grid-status/mcfarm/mcp10.15.01/runs.html • The job submission scripts exist, Anand our student converts them to DAGMAN • McView, the information provider (formerly known as CIA) has been released. It can be seen on page http://heppc1.uta.edu/atlas/grid-status/mcfarm/mcview.html

  9. Status ofMcView • It is inspired by GridView, a software tool by Kaushik De from UTA developed for Atlas Grid test bed (http://heppc1.uta.edu/atlas/grid-status/) • But McView takes the concept of Grid monitoring one step furhter: It reads not only information that is in MDS by deafult, but fills MDS with job status information as well. • This means: We have added new information providers for MDS.

  10. Status of McView – continued. • For the time being McView shows the number of undistributed jobs, errored jobs, running jobs and jobs ready to gather at each farm. It shows status of individual jobs as well • McView checks if the relevant daemons (monitor, gather, sam station) are alive at the farms • It detects stalled and inconsistent jobs and wars the operator.

  11. Conclusion • We would like to build a centrally operated, Globus based distributed system for MC production • It slowly starts to take shape. • It will be the first practical large scale implementation of Globus toolkit technology for HEP computing! • It will be a poor man’s Grid prototype, but nevertheless a first Grid-like computing network and a first step towards a real Grid!

More Related