Inspiration

gridvise is a holistic approach to large scale cluster management


The creators of gridvise have been involved in several large scale risk and regulatory reporting projects based on Oracle Coherence.  In each instance the clusters were managed through dozens of scripts taking care of cluster management - starting & stopping aspects of the cluster, rolling restart and other maintenance tasks.


In each case these scripts buried crucial and hard to maintain information about the cluster topology. With size and complexity this problem increased to a point where it affected decision making about changes to the cluster - people became reluctant to change service parameters, node counts, jvm and partition sizes because these adjustments were difficult to change consistently across all relevant scripts.


This was the inspriation for gridvise.


gridvise enables declarative cluster management.  There is one xml representation for the entire cluster - no matter how big it is - and by declaring a JVM it becomes enabled for all management aspects. It supports the concept of inheritance for JVM configuration, grouping and allocation to specific machines.


Download and try it yourself or view a demo here.