esxtop/resxtop

resxtop is lack of replay mode

reserved memory could be ballooned away.

CPU performance major counters:

• %RDY 10 Overpovisioning of vCPUs, execssive usage of vSMP, a limit has be set: %MLMTD
• %CSTP 3 Excessive usage of vSMP. Decrease amount of vCPUs for this particular VM, leads to increased scheduling opportunities.(ready/Co-deschedule State Time Percentage)
• %MLMTD 0 If larger than 0 the world is being throttled. Possible cause: limit on CPU (ResoucePool/World's limit setting)
• %SWPWT 5 (SWap Waiting Time) VM waiting on swapped pages to be read from disk: memory overcommitment.

memory performance major counters:

• MCTLSZ 1 If larger than 0, host is forcing VMs to inflate balloon driver to reclaim memory as host is overcommited.
• SWCU 1 if larger than 0, host has swapped memory pages in the past: over-commitment
• SWR/s 1 if larger than 0. host is actively reading from swap(vswap): excessive over-commitment
• SWW/s 1 if larger than 0. host is actively writing to swap(vswap): excessive over-commitment
• N%L 80 if less than 80, VM experience poor NUMA locality: ESX scheduler doesn't attempt to use NUMA optimization for the VM, and use memory via interconnect.(NRMEM)

network performance major counters:

• %DRPTX 1 Drop packets transmitted: very high  network utilization
• %DRPRX 1 Drop packet received: very high  network utilization

storage performance major counter:

• GAVG 25 = DAVG + KAVG
• DAVG 25 Disk latency most likely caused by array
• KAVG 2 Disk latency caused by VMkernel, means queuing, check "QUED"
• QUED 1 Queue maxed out. Possibly queue depth set to low, check with vendor for optimal queue depth value.
• ABRTS/s 1 Aborts issued by VM, because storage is not responding (60second for Windows): path failure or array not accepting I/O 
• RESETS/s 1 number of commands reset per second