r6 - 13 Jun 2011 - 15:53:21 - RobertGardnerYou are here: TWiki >  Admins Web > FacilitiesMeetingConfigVM

FacilitiesMeetingConfigVM

View the Indico agenda here: https://indico.cern.ch/event/vmcfg11.

Background

US ATLAS Facilities meeting June 15 & 16, 2011 at BNL (Berkner Room B) for site admins to share information about:

  • Virtual machine management strategies and experience gained from US ATLAS sites and elsewhere to date
  • Consolidation of head node services into virtual machines
  • Looking forward for configuration management approaches

Possible topics:

  • KVM, VMware, etc experiences in the Facility
  • Node and image provisioning with Cobbler, Rocks, etc
  • CernVM appliances
  • Configuration management with Puppet templates, Rocks, ...
  • Best practices with rpm and yum, repos
  • Head node service candidates: job schedulers, cluster monitoring, OSG grid services (eg. GUMS), storage services and databases
  • Benchmark case studies
  • Redundancy and failover
  • SL6 and KSM, http://www.linux-kvm.org/page/KSM - what can be tested with SL6 in advance of workshop? Memory footprint reduction.
  • KVM virtualization management tools: http://www.linux-kvm.org/page/Management_Tools
    • Which are best-of-breed?
    • Use case could be a high-availability server for critical services, such as condor master
    • demo with failover to second host
  • SANs for vm management

The meeting would be organized as one kick-off day of presentations and discussion followed by a second day of informal discussions of technical issues among participants.

Questions to guide discussions

Representatives from each site: please prepare a few slides addressing the following

  1. How are nodes built? Distinguish: worker nodes, head nodes, storage nodes, use of templates, as needed. Which "cluster services" are used (dhcp, dns, ..etc) and network configurations (use of VLANs, NAT, etc)
  2. How is configuration managed post-install?
  3. Describe update methods, eg., use of yum repos (which?)
  4. Describe virtualization frameworks in use, and which services are running in vm's.
  5. Describe any virtual cluster management frameworks in use (if any) and/or how are virtual machines administered at your site: instance creation, stop/start, migration and/or failover, etc.
  6. Describe any benchmarks studies of vm's

Agenda

  • First day - presentations from admins and other experts and discussion
  • Second day - informal technical discussions

Logistics

References


-- RobertGardner - 17 May 2011

About This Site

Please note that this site is a content mirror of the BNL US ATLAS TWiki. To edit the content of this page, click the Edit this page button at the top of the page and log in with your US ATLAS computing account name and password.


Attachments

 
Powered by TWiki
This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback