hsqueeze - Dynamic power management
hsqueeze {backend options...} [algorithm options...] [reporting options...]
hsqueeze –version
Backend options:
{ -L[ path ] [-X]** | -t data-file }
Algorithm options:
[ –minimal-resources=*factor* ] [ –target-resources=*factor* ]
Reporting options:
[ -S *file* ] [ -C[*file*] ]
hsqueeze does dynamic power management, by powering up or shutting down nodes, depending on the current load of the cluster. Currently, only suggesting nodes is implemented.
hsqueeze considers all online non-master nodes with only externally mirrored instances as candidates for being taken offline. These nodes are iteratively, starting from the node with the least number of instances, added to the set of nodes to be put offline, if possible. A set of nodes is considered as suitable for being taken offline, if, after marking these nodes as offline, balancing the cluster by the algorithm used by hbal(1) yields a situation where all instances are located on online nodes, and each node has at least the target resources free for new instances.
All offline nodes with a tag starting with htools:standby are considered candidates for being taken online. Those nodes are taken online till balancing the cluster by the algorithm used by hbal(1) yields a situation where each node has at least the minimal resources free for new instances.
-X | When using the Luxi backend, hsqueeze can also execute the given commands. The execution of the job series can be interrupted, see below for signal handling. |