Problem

High end, and midrange, systems can take a significant amount of time to collect a non-disruptive Hypervisor Resource Dump. If the purpose is to obtain a high level view of the placement, or confirm good placement, a Hypervisor Macro Resource Dump can be done, instead.

On fully populated High End systems, a Hypervisor Resource Dump may take several hours to collect, and occupy several hundred MB of space on the HMC. In comparison, Hypervisor Macro Resource Dumps generally run in seconds, and usually take less than 1 MB of space. The Hypervisor Macro Resource Dumps are usually much more scoped in nature, collecting a specific set of data.

The document titled, "How to Initiate a Resource Dump from the HMC" http://www.ibm.com/support/docview.wss?uid=nas8N1012174
describes how to collect the larger Hypervisor Resource Dump (selector 'system'), and also describes, generically, how to collect Hypervisor Macro Resource Dumps.  This document also details how to transmit the resulting dump file(s) to IBM.

Using the following selector string: hvlpconfigdata -base -procs -memory -domain -affinity
will generate a RSCDUMP type of dump file. Review the document linked above to see how the dump with a selector string may be initiated.
The following command may be issued at the HMC command line:
startdump -m {managed server} -t resource -r "hvlpconfigdata -base -procs -memory -domain -affinity"

It should be noted that this dump type will not indicate which processors in the system are licensed or not. If that information is needed, the larger, more complete Hypervisor Resource Dump with selector 'system' should be used.

These dump functions only apply to FW350, FW760, and later FW levels. It is recommended that the FW be at current fix level, to ensure the most accurate view of the system.