A PSOD is typically accompanied by a core dump file, which vm-support will collect when run (unless if you specified the -n switch.) Also, if you configure your server not to auto-reboot, then you can get a screen capture of the PSOD via a remote management device if available (e.g. an iLO on an HP server.)
If you're trying to analyze the core dump itself, I would think that's best left to VMware Support. But here's some links that may help depending on how much data you have:
http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=100425...
http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=100412... -- As this KB states, always get a screen capture of the PSOD. This typically means configuring your server to not auto-reboot if the hardware doesn't get a heartbeat from the OS within "X" minutes.
The stack trace in the core dump usually provides good clues as to the culprit of the PSOD. For example, it may be a NIC driver, or an HBA driver that threw an exception.
This blog post I found might give you what you want to know with trying to read a core dump yourself, though I've not tried this myself:
http://professionalvmware.com/2009/02/how-to-read-a-dump-esx-crash-dumps-that-is/
If you're trying to figure out where to start looking for log files, vmkwarning and vmkernel are usually the best ones to look through. But they may not have anything helpful if the PSOD came on suddenly.
