Port-xen archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
2nd round of xen benchmarking
I have swapped out one ancient system for another. This one has VMX
feature, but doesn't have the more modern virtualization features.
The basic benchmark is building devel/m4, no ccache, MAKE_JOBS=1, work
on tmpfs. 1 vcpu for dom0 and domU. (Multiple CPUs on bare metal, but
MAKE_JOBS=1.)
Overall, it looks like (talking about times) commpared to bare metal times:
dom0 is 2x
pv (or pvh/pvshim) domU is about 3x
n9-amd64: hvm is 19x (!!!!)
n10-amd64: pvh is 4.5 slower than pv, and pvhvm is 4x slower than pv
It remains a really good question if the newer CPUs will lead to reduced
slowdowns. As it is, pvh is a win for n10-i386, and other than that,
pv is best, sometimes by a huge amount.
I am really curious what others see.
----------------------------------------
* meta
** system config
dom0 and all domU have 1 vcpu
** _benchmark-pkg script
Builds have WRKDIR in tmpfs, because of mk.conf.
Builds are MAKE_JOBS=1, ignoring ccache.
4 runs of "time make package clean" in devel/m4.
Need to be careful about lockdebug! Runs below labeled LOCKDEBUG
(with a probability) are with all of DIAGNOSTIC, DEBUG, LOCKDEBUG.
Perhaps should be careful about branch, but for now all 2024Q2.
** disk
domU disks are generally 32 GB. On a domU, I measure disk read
performance (hvm, pv, and pv with a larger disk):
dd if=/dev/rwd0a of=/dev/null bs=1m
dd if=/dev/rxbd0a of=/dev/null bs=1m
dd if=/dev/rxbd0a of=/dev/null bs=1m count=32768
On the dom0, I measure the actual disk, and then the raw zvol for a
domU:
dd if=/dev/rwd0d of=/dev/null bs=1m count=32768
dd if=/dev/zvol/dsk/tank0/vm/n9-amd64 of=/dev/null bs=1m
* xen-Q6600
cpu0: "Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz"
cpu0: Intel Xeon 30xx, 32xx, 51xx, 53xx, 73xx, Core 2 Quad 6xxx, Core 2 Extreme 6xxx, Core 2 Duo 4xxx, 5xxx, 6xxx, 7xxx and Pentium DC (686-class)
cpu0: family 0x6 model 0xf stepping 0xb (id 0x6fb)
cpu0: features 0xbfebfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE>
cpu0: features 0xbfebfbff<MCA,CMOV,PAT,PSE36,CLFSH,DS,ACPI,MMX,FXSR,SSE,SSE2>
cpu0: features 0xbfebfbff<SS,HTT,TM,PBE>
cpu0: features1 0xe3bd<SSE3,DTES64,MONITOR,DS-CPL,VMX,EST,TM2,SSSE3,CX16,xTPR>
cpu0: features1 0xe3bd<PDCM>
cpu0: features2 0x20100800<SYSCALL/SYSRET,XD,EM64T>
cpu0: features3 0x1<LAHF>
cpu0: I-cache: 32KB 64B/line 8-way, D-cache: 32KB 64B/line 8-way
cpu0: L2 cache: 4MB 64B/line 16-way
cpu0: 64B prefetching
cpu0: ITLB: 128 4KB entries 4-way, 8 2M & 4 4M entries
cpu0: DTLB: 256 4KB entries 4-way, 16 4MB entries 4-way
https://www.cpubenchmark.net/cpu.php?cpu=Intel+Core2+Quad+Q6600+%40+2.40GHz&id=1038
CPUmark 1814 multi 951 single
** baremetal n10-amd64 8GB
81.82 real 55.51 user 24.08 sys
81.46 real 56.09 user 23.67 sys
81.43 real 55.28 user 24.47 sys
81.47 real 55.41 user 24.37 sys
wd0d: 34359738368 bytes transferred in 125.326 secs (274162890 bytes/sec)
zvol: 34359738368 bytes transferred in 239.504 secs (143462064 bytes/sec)
** baremetal n10-amd64 8GB LOCKDEBUG-probably
175.33 real 64.71 user 111.68 sys
173.85 real 64.71 user 110.74 sys
173.54 real 65.32 user 110.05 sys
173.50 real 64.99 user 110.39 sys
** dom0 n10-amd64 4GB
161.38 real 71.82 user 84.81 sys
160.98 real 72.16 user 84.75 sys
161.11 real 71.21 user 85.85 sys
161.15 real 72.44 user 84.63 sys
wd0d: 34359738368 bytes transferred in 139.977 secs (245467029 bytes/sec)
zvol: 34359738368 bytes transferred in 294.005 secs (116867870 bytes/sec)
** dom0 n10-amd64 4GB LOCKDEBUG-probably
289.38 real 74.93 user 206.64 sys
281.47 real 74.12 user 201.58 sys
283.08 real 74.60 user 203.00 sys
282.88 real 75.07 user 202.25 sys
** domU pvh/pvshim n9-i386 1GB
225.29 real 114.57 user 102.74 sys
231.95 real 119.89 user 105.01 sys
233.48 real 121.01 user 105.18 sys
230.41 real 118.90 user 104.48 sys
** domU pv n9-amd64 3GB
195.93 real 82.20 user 106.28 sys
195.34 real 81.62 user 107.12 sys
195.67 real 80.88 user 108.17 sys
195.36 real 82.74 user 106.34 sys
xbd0d: 34359738368 bytes transferred in 151.321 secs (227065234 bytes/sec)
** domU hvm n9-amd64 3GB
(manually mounted /dev/wd0a on /, because fstab has xbd0a)
1562.54 real 781.30 user 605.33 sys
1607.18 real 798.51 user 621.82 sys
1591.20 real 794.18 user 616.49 sys
1581.75 real 792.25 user 613.96 sys
wd0d: 34359738368 bytes transferred in 652.382 secs (52668127 bytes/sec)
** domU QEMU-noaccel/baremetal-dom0 n9-amd64 1GB
[works, slow, did not want to run benchmark!]
** domU pvh/pvshim n10-i386 1GB
221.40 real 116.13 user 98.72 sys
226.58 real 120.22 user 100.78 sys
225.94 real 118.69 user 101.83 sys
226.57 real 119.32 user 101.71 sys
** domU pvh n10-i386 1GB
161.77 real 90.56 user 66.04 sys
160.45 real 91.45 user 65.13 sys
160.60 real 90.65 user 66.03 sys
160.55 real 90.94 user 65.74 sys
** domU pv n10-amd64 3GB
207.45 real 86.78 user 118.86 sys
206.47 real 86.58 user 119.16 sys
206.80 real 86.81 user 119.18 sys
206.67 real 86.94 user 118.73 sys
# dd if=/dev/rxbd0a of=/dev/null bs=1m
xbd0d: 34359738368 bytes transferred in 187.584 secs (183169877 bytes/sec)
** domU pvh n10-amd64 3GB
852.86 real 470.35 user 354.56 sys
873.61 real 490.40 user 354.43 sys
884.74 real 494.53 user 363.39 sys
887.52 real 498.13 user 361.42 sys
xbd0d: 34359738368 bytes transferred in 204.111 secs (168338494 bytes/sec)
** domU pvhvm n10-amd64 3GB
815.92 real 446.63 user 343.02 sys
854.85 real 471.08 user 359.96 sys
858.88 real 469.04 user 366.19 sys
867.81 real 473.54 user 369.83 sys
xbd0d: 34359738368 bytes transferred in 200.763 secs (171145770 bytes/sec)
Home |
Main Index |
Thread Index |
Old Index