Hia,
just for fun I ran lmbench on Linux and IRIX, here are the results.
Some of the results are obviously bogus ...
The machines:
- Dull, a 200MHz Pentium MMX clone
- Indy, a 180MHz R5000SC machine
Ralf
L M B E N C H 1 . 9 S U M M A R Y
------------------------------------
(Alpha software, do not distribute)
Processor, Processes - times in microseconds - smaller is better
----------------------------------------------------------------
Host OS Mhz null null open selct sig sig fork exec sh
call I/O stat clos inst hndl proc proc proc
--------- ------------- ---- ---- ---- ---- ---- ----- ---- ---- ---- ---- ----
dull.coba Linux 2.0.27 200 1.1 1.8 29 44 0.08K 2.5 5 0.8K 5K 15K
indy.wald Linux 2.1.72 180 6.8 6.3 38 61 0.12K 6.3 18 2.0K 22K 70K
indy IRIX 6.2 180 2.9 10.0 88 110 2.44K 6.2 36 7.0K 19K 32K
^^^
Had this down to 1.8 for Linux 2.0.30, time to port the stuff to 2.1.
Context switching - times in microseconds - smaller is better
-------------------------------------------------------------
Host OS 2p/0K 2p/16K 2p/64K 8p/16K 8p/64K 16p/16K 16p/64K
ctxsw ctxsw ctxsw ctxsw ctxsw ctxsw ctxsw
--------- ------------- ----- ------ ------ ------ ------ ------- -------
dull.coba Linux 2.0.27 5 52 171 64 244 75 325
indy.wald Linux 2.1.72 4 251 155 655 155 655
indy IRIX 6.2 13 20 135 94 301 143 482
*Local* Communication latencies in microseconds - smaller is better
-------------------------------------------------------------------
Host OS 2p/0K Pipe AF UDP RPC/ TCP RPC/ TCP
ctxsw UNIX UDP TCP conn
--------- ------------- ----- ----- ---- ----- ----- ----- ----- ----
dull.coba Linux 2.0.27 5 19 36 89 212 139 289 1276
indy.wald Linux 2.1.72 4 34 52 103 280 165 406 597
indy IRIX 6.2 13 62 126 356 718 361 689 1321
File & VM system latencies in microseconds - smaller is better
--------------------------------------------------------------
Host OS 0K File 10K File Mmap Prot Page
Create Delete Create Delete Latency Fault Fault
--------- ------------- ------ ------ ------ ------ ------- ----- -----
dull.coba Linux 2.0.27 59 4 106 9 7582 2 0.1K
indy.wald Linux 2.1.72 14285 1
indy IRIX 6.2 450 813 1694 2325 3317 23.0K
*Local* Communication bandwidths in MB/s - bigger is better
-----------------------------------------------------------
Host OS Pipe AF TCP File Mmap Bcopy Bcopy Mem Mem
UNIX reread reread (libc) (hand) read write
--------- ------------- ---- ---- ---- ------ ------ ------ ------ ---- -----
dull.coba Linux 2.0.27 53 29 23 41 121 49 51 123 88
indy.wald Linux 2.1.72 55 24 6 26 60 0 0 0 0
indy IRIX 6.2 33 39 26 23 57 26 54 57 43
Memory latencies in nanoseconds - smaller is better
(WARNING - may not be correct, check graphs)
---------------------------------------------------
Host OS Mhz L1 $ L2 $ Main mem Guesses
--------- ------------- --- ---- ---- -------- -------
dull.coba Linux 2.0.27 200 10 88 146
indy.wald Linux 2.1.72 180 10 200 460
indy IRIX 6.2 180 11 197 485
|