Administrace komentářů

Dobrý den,

mám DELL server s řadičem H700 (02:00.0 RAID bus controller: Broadcom / LSI MegaRAID SAS 2108 [Liberator] (rev 05))

V něm mám zapojené 4x SATA 1TB 7k disky v RAID10 (VD0) a 4x SAS 600 15k v dalším RAID10 (VD1)

Zkusil jsem otestovat rychlost pomocí fio, a jestli dobře chapu ty údaje, tak VD0 (SATA) je znatelně rychlejší než VD1 (SAS) ?

Adapter 0 -- Virtual Drive Information:
Virtual Drive: 0 (Target Id: 0)
Name                :
RAID Level          : Primary-1, Secondary-0, RAID Level Qualifier-0
Size                : 1.818 TB
Sector Size         : 512
Mirror Data         : 1.818 TB
State               : Optimal
Strip Size          : 64 KB
Number Of Drives per span:2
Span Depth          : 2
Default Cache Policy: WriteBack, ReadAdaptive, Direct, No Write Cache if Bad BBU
Current Cache Policy: WriteBack, ReadAdaptive, Direct, No Write Cache if Bad BBU
Default Access Policy: Read/Write
Current Access Policy: Read/Write
Disk Cache Policy   : Enabled
Encryption Type     : None
Bad Blocks Exist: No
Is VD Cached: Yes
Cache Cade Type : Read Only


Virtual Drive: 1 (Target Id: 1)
Name                :
RAID Level          : Primary-1, Secondary-0, RAID Level Qualifier-0
Size                : 1.089 TB
Sector Size         : 512
Mirror Data         : 1.089 TB
State               : Optimal
Strip Size          : 64 KB
Number Of Drives per span:2
Span Depth          : 2
Default Cache Policy: WriteBack, ReadAdaptive, Direct, No Write Cache if Bad BBU
Current Cache Policy: WriteBack, ReadAdaptive, Direct, No Write Cache if Bad BBU
Default Access Policy: Read/Write
Current Access Policy: Read/Write
Disk Cache Policy   : Enabled
Encryption Type     : None
Bad Blocks Exist: No
Is VD Cached: Yes
Cache Cade Type : Read Only

FIO spouštím takto:

fio --name=random-write --ioengine=posixaio --rw=randwrite --bs=4k --size=4g --numjobs=1 --iodepth=1 --runtime=60 --time_based --end_fsync=1

A zde jsou vysledky:

Pro SATA pool:

random-write: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=1
fio-3.36
Starting 1 process
Jobs: 1 (f=1): [F(1)][100.0%][eta 00m:00s]                           
random-write: (groupid=0, jobs=1): err= 0: pid=7601: Wed Jan 15 19:52:39 2025
  write: IOPS=39.4k, BW=154MiB/s (161MB/s)(13.0GiB/86370msec); 0 zone resets
    slat (usec): min=3, max=34451, avg= 4.06, stdev=18.90
    clat (nsec): min=754, max=474798, avg=852.46, stdev=342.24
     lat (usec): min=4, max=34457, avg= 4.91, stdev=18.91
    clat percentiles (nsec):
     |  1.00th=[  788],  5.00th=[  804], 10.00th=[  812], 20.00th=[  820],
     | 30.00th=[  828], 40.00th=[  836], 50.00th=[  836], 60.00th=[  844],
     | 70.00th=[  852], 80.00th=[  868], 90.00th=[  884], 95.00th=[  900],
     | 99.00th=[ 1112], 99.50th=[ 1272], 99.90th=[ 3920], 99.95th=[ 5600],
     | 99.99th=[ 6752]
   bw (  KiB/s): min=11104, max=728024, per=100.00%, avg=681828.72, stdev=118657.92, samples=39
   iops        : min= 2776, max=182006, avg=170457.23, stdev=29664.50, samples=39
  lat (nsec)   : 1000=98.56%
  lat (usec)   : 2=1.32%, 4=0.03%, 10=0.09%, 20=0.01%, 50=0.01%
  lat (usec)   : 100=0.01%, 250=0.01%, 500=0.01%
  cpu          : usr=6.05%, sys=23.29%, ctx=43488, majf=0, minf=15
  IO depths    : 1=100.0%, 2=0.0%, 4=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, >=64=0.0%
     submit    : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
     complete  : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
     issued rwts: total=0,3405018,0,0 short=0,0,0,0 dropped=0,0,0,0
     latency   : target=0, window=0, percentile=100.00%, depth=1

Run status group 0 (all jobs):
  WRITE: bw=154MiB/s (161MB/s), 154MiB/s-154MiB/s (161MB/s-161MB/s), io=13.0GiB (13.9GB), run=86370-86370msec

Disk stats (read/write):
  sdb: ios=0/336221, sectors=0/27223840, merge=0/1328, ticks=0/4480718, in_queue=4480718, util=74.62%

a pro SAS pool:

random-write: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=libaio, iodepth=1
fio-3.36
Starting 1 process
Jobs: 1 (f=1): [w(1)][100.0%][eta 00m:00s]                         
random-write: (groupid=0, jobs=1): err= 0: pid=7629: Wed Jan 15 19:54:02 2025
  write: IOPS=27.3k, BW=107MiB/s (112MB/s)(8192MiB/76915msec); 0 zone resets
    slat (usec): min=3, max=171445, avg= 4.06, stdev=155.97
    clat (nsec): min=756, max=146517, avg=823.91, stdev=200.90
     lat (usec): min=4, max=171452, avg= 4.89, stdev=155.98
    clat percentiles (nsec):
     |  1.00th=[  780],  5.00th=[  788], 10.00th=[  796], 20.00th=[  796],
     | 30.00th=[  804], 40.00th=[  812], 50.00th=[  812], 60.00th=[  820],
     | 70.00th=[  820], 80.00th=[  828], 90.00th=[  844], 95.00th=[  868],
     | 99.00th=[  972], 99.50th=[ 1064], 99.90th=[ 1768], 99.95th=[ 5600],
     | 99.99th=[ 6304]
   bw (  KiB/s): min=128368, max=745832, per=100.00%, avg=671088.64, stdev=148465.89, samples=25
   iops        : min=32092, max=186458, avg=167772.16, stdev=37116.47, samples=25
  lat (nsec)   : 1000=99.20%
  lat (usec)   : 2=0.70%, 4=0.01%, 10=0.08%, 20=0.01%, 50=0.01%
  lat (usec)   : 250=0.01%
  cpu          : usr=4.23%, sys=15.52%, ctx=33897, majf=0, minf=25
  IO depths    : 1=100.0%, 2=0.0%, 4=0.0%, 8=0.0%, 16=0.0%, 32=0.0%, >=64=0.0%
     submit    : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
     complete  : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
     issued rwts: total=0,2097153,0,0 short=0,0,0,0 dropped=0,0,0,0
     latency   : target=0, window=0, percentile=100.00%, depth=1

Run status group 0 (all jobs):
  WRITE: bw=107MiB/s (112MB/s), 107MiB/s-107MiB/s (112MB/s-112MB/s), io=8192MiB (8590MB), run=76915-76915msec

Disk stats (read/write):
  sdc: ios=0/100982, sectors=0/16777328, merge=0/1226, ticks=0/2752731, in_queue=2752732, util=83.47%

Čekal bych že 15k SAS by mohl být znatelně rychlejší než SATA, ale ne že to bude naopak.Oba pooly maji stejné nastavení, běží na jednom serveru a řadiči. Procházím si parametry v MegaCli, ale nejaký zjevný problém nevidím. (ale hlavnš se s tím učím, takže tam možná bude nějaká drobná ale zásadní "banalita" )

Nemáte nejaký nápad, proč to tak je? Co zkontrolovat?

Děkuji.