XO battery/performance
Yioryos Asprobounitis
mavrothal at yahoo.com
Wed May 30 03:34:43 EDT 2012
> If you want an idea of low-level performance, I can suggest
> running LMBench.
>
Got the Debian lmbench_3.0-a7 source that compiles and runs fine w/o bitkeeper.
Run the hardware part of the tests on the XO-1.5 (os880) and xo-1.75 (os12- correct kernel) with the same configuration.
What was striking was that the XO-1.75 used 25% of the battery for 1 run while the XO-1.5 used 65% of the battery!
Most of the test had empty values but the informative ones (below) show that the XO-1.5 is better in basic integer operations and memory bandwidth while the XO-1.75 is better in float and double operations as well as in memory latency.
I'm not sure how much this means for real life usage :-/
But since I did it, here are the results and a comparison based on the best of 3 values for each machine (hopefully the text alignment will be preserved).
L M B E N C H 3 . 0 S U M M A R Y
------------------------------------
(Alpha software, do not distribute)
Basic system parameters
------------------------------------------------------------------------------
Host OS Description Mhz tlb cache mem scal
pages line par load
bytes
--------- ------------- ----------------------- ---- ----- ----- ------ ----
xo-1.5 Linux 2.6.35. i686-pc-linux-gnu 1000 64 64 2.3900 1
xo-1.75 Linux 3.0.19_ armv7l-linux-gnu 796 8 32 1.0000 1
Basic integer operations - times in nanoseconds - smaller is better
-------------------------------------------------------------------
Host OS intgr intgr intgr intgr intgr
bit add mul div mod
--------- ------------- ------ ------ ------ ------ ------
xo-1.5 Linux 2.6.35. 1.0100 0.0400 1.3500 96.5 55.2
xo-1.75 Linux 3.0.19_ 1.2600 0.1400 3.8800 153.5 33.8
1.5 > 1.75 1.2475 3.5 2.874 1.590 0.612
Basic float operations - times in nanoseconds - smaller is better
-----------------------------------------------------------------
Host OS float float float float
add mul div bogo
--------- ------------- ------ ------ ------ ------
xo-1.5 Linux 2.6.35. 7.0300 7.5400 73.3 92.5
xo-1.75 Linux 3.0.19_ 6.2800 7.5400 26.4 50.5
1.75 > 1.5 1.119 1.000 2.777 1.832
Basic double operations - times in nanoseconds - smaller is better
------------------------------------------------------------------
Host OS double double double double
add mul div bogo
--------- ------------- ------ ------ ------ ------
xo-1.5 Linux 2.6.35. 7.0300 8.0400 73.5 92.5
xo-1.75 Linux 3.0.19_ 6.2800 7.5400 44.0 76.9
1.75 > 1.5 1.119 1.066 1.670 1.203
*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
--------- ------------- ---- ---- ---- ------ ------ ------ ------ ---- -----
xo-1.5 Linux 2.6.35. 306.1 307.0 458. 646.6
xo-1.75 Linux 3.0.19_ 151.0 143.6 216. 328.1
1.5 > 1.75 2.02 2.13 2.12 1.97
Memory latencies in nanoseconds - smaller is better
(WARNING - may not be correct, check graphs)
------------------------------------------------------------------------------
Host OS Mhz L1 $ L2 $ Main mem Rand mem Guesses
--------- ------------- --- ---- ---- -------- -------- -------
xo-1.5 Linux 2.6.35. 1000 6.0720 29.1 100.7 361.7
xo-1.75 Linux 3.0.19_ 796 3.8060 19.3 99.8 257.7
1.75 > 1.5 1.595 1.508 1.009 1.404
More information about the Devel
mailing list