Navigation and service


Graphical monitoring of batch system controlled cluster

-> development release version 1.41 now available
-> development release version 1.21 of LLview data access component now available
(see Release notes)

LLview is a client-server based application which allows to monitor the utilization of clusters controlled by batch systems like IBM LoadLeveler, PBSpro, Torque, or IBM Blue Gene system data base. On large supercomputer clusters with several thousands of processors it is not feasible to monitor the usage of system and batch load with command line tools, because the lists or tables in their output are becoming too large and complex. Important information for example is the usage of nodes/processors and the required resources of running and waiting jobs. LLview gives a quick and compact summary of this information.

LLview Snapshot
Main window of LLview

Graphical elements of LLview are a node display, a usage bar, which gives a direct view of the job-granularity, a list of running jobs and a list of waiting jobs, and a graph chart displaying the number of jobs in the different queues. A very flexible type of charts for displaying job profiles for I/O, memory and CPU usage over time is introduced with client version 1.41. Besides, generic annotations are integrated, which allow for arbitrarily placed texts and images to enrich LLview displays with additional information on any location of the screen.

Further information:
-> Components and data flow of LLview
-> Using the LLview client
-> Managing Configuration Options
-> Copyright and Disclaimer
-> Download and Installation
-> LML_da and its configuration
-> Additional tools, adaptors for batch systems
-> Introduction slides