ESXi Performance: A microscopic approach

sched-stats -t numa-global

sched-stats -t numa-pnode

sched-stats -t numa-clients

sched-stats -t numa-migration
vsish -e get /hardware/cpu/cpuModelName
vsish -e cat /sched/globalStats/memory/memStats

vsish -e get /config/Numa/intOpts/LTermFairnessInterval

vsish -e cat /sched/numaStats/numaNodes/0
vsish -e cat /sched/numaStats/numaNodes/1
vsish -e cat /sched/globalStats/memory/memStats | grep memory-state

vmdumper -l | cut -d \/ -f 2-5 | while read path; do egrep -oi "DICT.*(displayname.* | numa.* | cores.* | vcpu.* | memsize.* | affinity.*) = .*lnuma:.* | numaHost:.*" "/$path/vmware.log"; echo -e; done
memstats -r comp-stats -s

memstats -r comp-stats -s total:minfree:free:managedByMemSched -u mb
memstats -r comp-stats -s total:kernelCode:dataAndHeap:buddyOvhd:vmkClientConsumed -u mb

memstats -r vm-stats -s name:memSize:allocTgt:mapped:consumed:touched -u mb
esxcli hardware platform get
esxcfg-scsidevs --list
esxcfg-scsidevs --vmfs
esxcfg-mpath --list-paths

esxcli network nic tso get

esxcli system settings advanced list -o /Net/UseHwTSO
esxcli vsan policy getdefault
esxcli storage vmfs extent list
esxcli system module parameters list -m elxnet
esxcli system settings advanced list -o /Net/TcpipDefLROEnabled
esxcli system settings advanced list -o /Net/VmxnetLROMaxLength
esxcli system settings advanced list -o /Net/Vmxnet3HwLRO
esxcli system settings advanced list -o /Net/Vmxnet3SwLRO

esxcli network nic cso get

esxcli system module parameters list -m bnx2x

esxcli network nic coalesce get
esxcli network nic ring current get -n vmnic0
esxcli network nic ring current get -n vmnic1

vmkchdev -l | grep vmnic

Try these out (learnt from Mike on 19th Aug 2019)
--------------------------------------------------------------
esxcfg-nas -l
cat /etc/vmware/hostd/vmInventory.xml
vmkload_mod -s nmlx5_core | grep Version
vmkchdev -l | grep vmnic0
vim-cmd vimsvc/task_list
vsish -e ls /net/pNics | while read nics; do echo -n $nics; vsish -e cat /net/pNics/${nics}stats; done | less
vmkbacktrace -n hostd -c -w 
esxtop -a -b -d 5 -n 360 
vm-support -w .

1.     Obtain a live core from hostd during the problem

2.     Generate an ESXi Host log bundle after the live core, and during the problem:
vm-support -w /vmfs/volumes/DATASTORE/

3.     Intentionally crash the Host, generating a core dump all on Host memory:
vsish -e set /reliability/crashMe/Panic 1
/> CrashMe