Understanding the evolution of conditions data access through Frontier for the ATLAS Experiment
- Czech Academy of Sciences, Prague (Czech Republic)
- Sapienza Univ. of Rome (Italy)
- Science and Technology Facilities Council (STFC), Daresbury (United Kingdom)
- National Centre for Scientific Research (CNRS), Paris (France)
- Univ. of Valencia (Spain)
- Univ. of Texas, Arlington, TX (United States)
- Fermi National Accelerator Lab. (FNAL), Batavia, IL (United States)
The ATLAS Distributed Computing system uses the Frontier system to access the Conditions, Trigger, and Geometry database data stored in the Oracle Offline Database at CERN by means of the HTTP protocol. All ATLAS computing sites use Squid web proxies to cache the data, greatly reducing the load on the Frontier servers and the databases. One feature of the Frontier client is that in the event of failure, it retries with different services. While this allows transient errors and scheduled maintenance to happen transparently, it does open the system up to cascading failures if the load is high enough. Throughout LHC Run 2 there has been an ever increasing demand on the Frontier service. There have been multiple incidents where parts of the service failed due to high load. A significant improvement in the monitoring of the Frontier service was required. The monitoring was needed to identify both problematic tasks, which could then be killed or throttled, and to identify failing site services as the consequence of a cascading failure is much higher. This presentation describes the implementation and features of the monitoring system.
- Research Organization:
- Fermi National Accelerator Lab. (FNAL), Batavia, IL (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC), High Energy Physics (HEP)
- Contributing Organization:
- ATLAS Collaboration
- Grant/Contract Number:
- AC02-07CH11359
- OSTI ID:
- 1570194
- Report Number(s):
- FERMILAB-CONF-18-583-CD; oai:inspirehep.net:1758380; TRN: US2001153
- Journal Information:
- EPJ Web of Conferences, Vol. 214; Conference: 23rd International Conference on Computing in High Energy and Nuclear Physics, Sofia, Bulgaria, 07/09-07/13/2018; ISSN 2100-014X
- Publisher:
- EDP SciencesCopyright Statement
- Country of Publication:
- United States
- Language:
- English
Greatly improved cache update times for conditions data with Frontier/Squid
|
journal | April 2010 |
CMS conditions data access using FroNTier
|
journal | July 2008 |
Similar Records
Operational Experience with the Frontier System in CMS
Greatly improved cache update times for conditions data with Frontier/Squid