Speaker
Description
URL for further information
sls.cern.ch , http://sls.cern.ch/sls/service.php?id=ServicesForATLAS ...
Keywords
service level status, monitoring, availability
Detailed analysis
The LHC experiments are using an increasingly number of complex and heterogeneous services: SLS is a framework that allows to group all these different services, and to report their status and their availability.
The SLS system addresses these needs by providing a web-based display. It dynamically shows availability, basic information and statistics about these services, as well as the dependencies between them. The SLS framework has been developed by the CERN-IT/FIO group and has been adopted to monitor different services:
SLS is currently dealing with more than 350 services, spawning from administrative applications, to physics and infrastructure services, from Grid related to experiment specific services.
Justification for delivering demo and technical requirements (ONLY for demonstrations)
A demonstrations could be useful to explain the different functionality for shifters and experts, allowing them to interact lively with the SLS framework.
Impact
The modularity of SLS allows providing different views for different end-users. The various services can provide, together with the mandatory availability status, many parameters to SLS via an user-friendly format. SLS can easily represent complex services, which are composed by different subservices. For each service thresholds can be defined to trigger alarms of increasing severity. Historical data available in the SLS database can be retrieved via the web interface. The SLS monitoring information can be easily accessed via a programmatic interface and displayed via other visualization tools, like GridMap.
SLS is now effectively used to monitor the status of the LHC experiments service.
Conclusions and Future Work
SLS is now effectively used to monitor the status of the LHC experiments service. Many more Grid or experiment specific services can be integrated into the SLS framework, and work is in progress to have a scalable system that will provide a global view of the LHC Distributed Computing infrastructure.
