1 / 15

Tier-1 Evolution and Futures

Tier-1 Evolution and Futures. GridPP 29, Oxford Ian Collier September 27 th 2012. Where are we going?. Where are we going?. Good question. Where are we going?. Good question We know things are changing. Where are we going?. Good question We know things are changing

eyal
Télécharger la présentation

Tier-1 Evolution and Futures

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. Tier-1 Evolution and Futures GridPP 29, Oxford Ian Collier September 27th 2012

  2. Where are we going?

  3. Where are we going? • Good question

  4. Where are we going? • Good question • We know things are changing

  5. Where are we going? • Good question • We know things are changing • We don’t know how things will turn out

  6. Where are we going? • Good question • We know things are changing • We don’t know how things will turn out • We can make choices

  7. Where are we going? • Good question • We know things are changing • We don’t know how things will turn out • We can make choices • We must be ready to try things and to adapt

  8. Where we are • Services running. • Lots of cores • Lost of disk • Lots of tape • Castor running well & stably • Shaun will talk about storage future • EMI/UMD Middleware rolling out • Continues to be a moving target • Virtualisation • Usage growing fast • Configuration management • Batch & disk farms much more tractable • Deploying new instances of services easier

  9. 2012 • Virtualisation • Heading to most services running on resilient virtualisation platform with shared storage • Unless there is a definite contra-indication (performance) • Currently MS Hyper-V • It works. But it is Windows (unfamiliar territory still) • Likely to migrate to open source solution (ovirt/RHEV) • Cloud • Stratuslabprototype • Continuing EMI/UMD rollout • Focus on EMI-2 • CVMFS Stratum 0 for non-LHC VOs • Ready for testing by VOs

  10. Virtualised hosts Fabric - Tier1 Review, June 2012

  11. 2012 • Combination of Hyper-V and SCD cloud • Deploying new services easier • Testing things much much easier • Combined with Quattor (fabric & configuration management) it is pretty easy to roll out, for example, a test cluster

  12. Getting nebulous • Tier 1 developing SCD Cloud platform • Using StratusLab – integrating cloud & grid • Very much a prototype so far • Several dozen VMs there as well • Very good for ‘just trying things’ • Several use cases: • Internal (Tier1 & SCT) development & testbeds • At present trust users as regular sysadmins • Scientific computing resource for other departments in STFC • Virtualised layer in batch farm • Experiments are investigating clouds • We need to be building up experience • EGI Federated cloud

  13. Looking to Future • Integrating Aquilon • ‘next generation’ configuration database for Quattor • SQL database plus git workflow • Faster more flexible development & deployment • Better relational integrity for the ‘database like’ parts • STFC deploying new DNS management • We will have direct access to manage our networks including API • More dynamic provisioning possible • Entirely more ‘dynamic’ infrastructure? • See cloud work…

  14. Looking to Future • There are emerging themes • Virtualisationmaking hardware provisioning easier • Can abstract hardware delivery/maintenance from service provision • As site admins we like this • Cloud frameworks should allow • containerisation • Users may like this • Within some higher level requirements set by sites users can have much more choice/control of their software environments • More elastic & dynamic access to resources • Cloud interfaces may make access to resources easierfor users

  15. Where are we going? • We know things are changing • We don’t know how it will turn out • We can make choices • Have to be ready to try things and to adapt

More Related