Reasonable and Expected Load Averages
Hi. I'm just curious as to what i should expect from my new hardware. I'll put the ouput from lscpu below. There are about 100 low to medium traffic sites on the server, most with a couple low volume mail accounts. The home directories are on an array of hybrid drives, the OS is on SSD drives. There is 256mb ram. What would be considered good load averages when running top?
[verdon@host2 ~]$ lscpu
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 96
On-line CPU(s) list: 0-95
Thread(s) per core: 2
Core(s) per socket: 24
Socket(s): 2
NUMA node(s): 2
Vendor ID: AuthenticAMD
CPU family: 23
Model: 49
Model name: AMD EPYC 7402 24-Core Processor
Stepping: 0
CPU MHz: 2800.000
CPU max MHz: 2800.0000
CPU min MHz: 1500.0000
BogoMIPS: 5589.77
Virtualization: AMD-V
L1d cache: 32K
L1i cache: 32K
L2 cache: 512K
L3 cache: 16384K
NUMA node0 CPU(s): 0-23,48-71
NUMA node1 CPU(s): 24-47,72-95
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht
syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf pni pclmu
lqdq monitor ssse3 fma cx16 sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy
abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb
cat_l3 cdp_l3 hw_pstate ssbd mba ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt
clwb sha_ni xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr wbn
oinvd amd_ppin arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avi
c v_vmsave_vmload vgif v_spec_ctrl umip r
-
Hey there! It's really impossible to say - one busy site could take down a server and cause serious load, or 100 static sites with no traffic could have almost no load on the machine.
The best answer is the monitor the server at various times over several days to get a baseline of how the system is performing.
0 -
@cPRex Thanks :-) I understand and appreciate that. There must be some general rule of thumb though… a 15 minute load average of 1.0, 4.0, 10.0, 20.0… when does it become concerning? I believe there's a formula based on # of cpu that should give you a rough idea.
0 -
A load of 1 on a server with 1 CPU core means that single core is maxed out, so that isn't ideal. In your case, if you have 96 cores, you could easily have a load of 100, but that seems unlikely to happen.
0 -
That’s what I thought, but I start getting antsy with loads over 5… I’d be calling the fire dept if my load got anywhere near 100! Thanks for your input. It's very much appreciated :-)
0 -
Hahaha - that's something we all have to get used to with these modern multi-core CPUs - a load of 10 could really be something like 0.1!
1 -
@abu islam
Thank you for the thorough and thoughtful reply. Good catch on the RAM… yes, GB.0
Please sign in to leave a comment.
Comments
6 comments