As a best practice, you should consider using a centralized system for all logs in your infrastructure to cover all elements, instead of only using VMware vSphere to correlate events.

For example, let's say that at 17:01:03, your ESXi hypervisor loses half of the active paths to the storage array. If you have a centralized log management system in place, you can easily search the logs, and you might find that at the same time, your data center switched to the UPS system because of a power failure, but one of the switches went down (the power supply in the switch was not connected to the UPS system).

There are several products you can use for this kind of task. Some are free, and others aren't.