AIM9 results

Benchmark Paolo Ciarrocchi 5 participants 8 messages Dec 16, 2002 – Dec 23, 2002

lore

Share & tools

lore

Paolo Ciarrocchi

author

2002-12-16 22:45 UTC #0 lore raw

Hi all,
hw is an Omnibook 6000 (laptop)
fs is Reiserfs

format:
2.4.19
2.5.51
2.5.52

add_double 10030 23.9282       430707.88 Thousand Double Precision Additions/second
add_double 10030 23.7288       427118.64 Thousand Double Precision Additions/second
add_double 10030 23.7288       427118.64 Thousand Double Precision Additions/second

add_float 10000 35.9 430800.00 Thousand Single Precision Additions/second
add_float 10010 35.5644 426773.23 Thousand Single Precision Additions/second
add_float 10010 35.5644 426773.23 Thousand Single Precision Additions/second

add_long 10010 22.1778      1330669.33 Thousand Long Integer Additions/second
add_long 10020 21.9561      1317365.27 Thousand Long Integer Additions/second
add_long 10020 21.9561      1317365.27 Thousand Long Integer Additions/second

add_int 10020 22.0559      1323353.29 Thousand Integer Additions/second
add_int 10020 21.8563      1311377.25 Thousand Integer Additions/second
add_int 10020 21.8563      1311377.25 Thousand Integer Additions/second

add_short 10000 55.4      1329600.00 Thousand Short Integer Additions/second
add_short 10000 54.9      1317600.00 Thousand Short Integer Additions/second
add_short 10000 54.9      1317600.00 Thousand Short Integer Additions/second

creat-clo 10040 19.4223 19422.31 File Creations and Closes/second
creat-clo 10010 87.2128 87212.79 File Creations and Closes/second
creat-clo 10000 75.3 75300.00 File Creations and Closes/second
^^^Here 2.5.* is _a lot_ faster than 2.4.19 but 2.5.52 is slower then 2.5.51

page_test 10000 123.9 210630.00 System Allocations & Pages/second
page_test 10000 105.7 179690.00 System Allocations & Pages/second
page_test 10010 105.095 178661.34 System Allocations & Pages/second
^^^^ 2.5.5* is still slower then 2.4.19

brk_test 10010 48.951 832167.83 System Memory Allocations/second
brk_test 10020 44.6108 758383.23 System Memory Allocations/second
brk_test 10010 43.8561 745554.45 System Memory Allocations/second
^^^^ 2.5.5* is still slower then 2.4.19

jmp_test 10000 4313.7      4313700.00 Non-local gotos/second
jmp_test 10000 4273.2      4273200.00 Non-local gotos/second
jmp_test 10000 4273.1      4273100.00 Non-local gotos/second

signal_test 10000 166.1       166100.00 Signal Traps/second
signal_test 10000 157.4       157400.00 Signal Traps/second
signal_test 10000 159.6       159600.00 Signal Traps/second

^^^^ 2.5.5* is still slower then 2.4.19

exec_test 10000 13.8 69.00 Program Loads/second
exec_test 10020 12.9741 64.87 Program Loads/second
exec_test 10030 12.9611 64.81 Program Loads/second
^^^^ 2.5.5* is still slower then 2.4.19

fork_test 10000 44.8 4480.00 Task Creations/second
fork_test 10020 23.9521 2395.21 Task Creations/second
fork_test 10020 29.0419 2904.19 Task Creations/second
^^^^ 2.5.5* is still slower then 2.4.19

link_test 10000 155.3 9783.90 Link/Unlink Pairs/second
link_test 10000 147.7 9305.10 Link/Unlink Pairs/second
link_test 10010 140.959 8880.42 Link/Unlink Pairs/second
^^^^ 2.5.5* is still slower then 2.4.19

disk_rr 10050 6.96517        35661.69 Random Disk Reads (K)/second
disk_rr 10110 7.51731        38488.63 Random Disk Reads (K)/second
disk_rr 10050 7.46269        38208.96 Random Disk Reads (K)/second

^^^^ 2.5.5* is faster then 2.4.19

disk_rw 10060 5.666 29009.94 Random Disk Writes (K)/second
disk_rw 10000 6.8 34816.00 Random Disk Writes (K)/second
disk_rw 10090 6.73935 34505.45 Random Disk Writes (K)/second
^^^^ 2.5.5* is faster then 2.4.19

disk_rd 10010 38.1618 195388.61 Sequential Disk Reads (K)/second
disk_rd 10020 38.024 194682.63 Sequential Disk Reads (K)/second
disk_rd 10010 38.0619 194877.12 Sequential Disk Reads (K)/second

disk_wrt 10100 8.51485 43596.04 Sequential Disk Writes (K)/second
disk_wrt 10070 9.43396 48301.89 Sequential Disk Writes (K)/second
disk_wrt 10080 9.3254 47746.03 Sequential Disk Writes (K)/second
^^^^ 2.5.5* is faster then 2.4.19

disk_cp 10120 7.31225        37438.74 Disk Copies (K)/second
disk_cp 10010 7.69231        39384.62 Disk Copies (K)/second
disk_cp 10090 7.63132        39072.35 Disk Copies (K)/second

^^^^ 2.5.5* is faster then 2.4.19

sync_disk_rw 16020 0.062422 159.80 Sync Random Disk Writes (K)/second
sync_disk_rw 18790 0.0532198 136.24 Sync Random Disk Writes (K)/second
sync_disk_rw 14570 0.0686342 175.70 Sync Random Disk Writes (K)/second
^^^^ 2.5.5* is faster then 2.4.19

sync_disk_wrt 11220 0.0891266          228.16 Sync Sequential Disk Writes (K)/second
sync_disk_wrt 10150 0.0985222          252.22 Sync Sequential Disk Writes (K)/second
sync_disk_wrt 10130 0.0987167          252.71 Sync Sequential Disk Writes (K)/second

^^^^ 2.5.5* is faster then 2.4.19

sync_disk_cp 10870 0.0919963 235.51 Sync Disk Copies (K)/second
sync_disk_cp 19580 0.102145 261.49 Sync Disk Copies (K)/second
sync_disk_cp 10050 0.0995025 254.73 Sync Disk Copies (K)/second
^^^^ 2.5.5* is faster then 2.4.19

disk_src 10000 118.2 8865.00 Directory Searches/second
disk_src 10010 110.889 8316.68 Directory Searches/second
disk_src 10010 123.477 9260.74 Directory Searches/second
^^^^ 2.5.5* is faster then 2.4.19 and 2.5.52 is faster then 2.5.51

div_double 10000 24.4 73200.00 Thousand Double Precision Divides/second
div_double 10020 24.1517 72455.09 Thousand Double Precision Divides/second
div_double 10010 24.1758 72527.47 Thousand Double Precision Divides/second

div_float 10010 24.3756        73126.87 Thousand Single Precision Divides/second
div_float 10010 24.1758        72527.47 Thousand Single Precision Divides/second
div_float 10010 24.1758        72527.47 Thousand Single Precision Divides/second

div_long 10020 19.9601        17964.07 Thousand Long Integer Divides/second
div_long 10020 19.7605        17784.43 Thousand Long Integer Divides/second
div_long 10010 19.7802        17802.20 Thousand Long Integer Divides/second

div_int 10020 19.9601        17964.07 Thousand Integer Divides/second
div_int 10010 19.7802        17802.20 Thousand Integer Divides/second
div_int 10020 19.7605        17784.43 Thousand Integer Divides/second

div_short 10030 19.9402        17946.16 Thousand Short Integer Divides/second
div_short 10020 19.7605        17784.43 Thousand Short Integer Divides/second
div_short 10010 19.7802        17802.20 Thousand Short Integer Divides/second

fun_cal 10000 62.5     32000000.00 Function Calls (no arguments)/second
fun_cal 10000 61.9     31692800.00 Function Calls (no arguments)/second
fun_cal 10000 61.9     31692800.00 Function Calls (no arguments)/second

fun_cal1 10000 170.4 87244800.00 Function Calls (1 argument)/second
fun_cal1 10010 168.831 86441558.44 Function Calls (1 argument)/second
fun_cal1 10010 168.831 86441558.44 Function Calls (1 argument)/second

fun_cal2 10000 112.5 57600000.00 Function Calls (2 arguments)/second
fun_cal2 10000 111.5 57088000.00 Function Calls (2 arguments)/second
fun_cal2 10010 111.389 57030969.03 Function Calls (2 arguments)/second

fun_cal15 10010 34.0659     17441758.24 Function Calls (15 arguments)/second
fun_cal15 10010 33.7662     17288311.69 Function Calls (15 arguments)/second
fun_cal15 10010 33.7662     17288311.69 Function Calls (15 arguments)/second

sieve 10450 0.861244            4.31 Integer Sieves/second
sieve 10600 0.849057            4.25 Integer Sieves/second
sieve 10600 0.849057            4.25 Integer Sieves/second

mul_double 10020 21.5569 258682.63 Thousand Double Precision Multiplies/second
mul_double 10020 21.3573 256287.43 Thousand Double Precision Multiplies/second
mul_double 10030 21.336 256031.90 Thousand Double Precision Multiplies/second

mul_float 10030 21.5354 258424.73 Thousand Single Precision Multiplies/second
mul_float 10030 21.336 256031.90 Thousand Single Precision Multiplies/second
mul_float 10020 21.3573 256287.43 Thousand Single Precision Multiplies/second

mul_long 10000 947.7       227448.00 Thousand Long Integer Multiplies/second
mul_long 10000 939.6       225504.00 Thousand Long Integer Multiplies/second
mul_long 10000 939.8       225552.00 Thousand Long Integer Multiplies/second

mul_int 10000 951.9       228456.00 Thousand Integer Multiplies/second
mul_int 10000 943.2       226368.00 Thousand Integer Multiplies/second
mul_int 10000 943.3       226392.00 Thousand Integer Multiplies/second

mul_short 10000 759.1       227730.00 Thousand Short Integer Multiplies/second
mul_short 10000 753.4       226020.00 Thousand Short Integer Multiplies/second
mul_short 10000 753.4       226020.00 Thousand Short Integer Multiplies/second

num_rtns_1 10000 467.2        46720.00 Numeric Functions/second
num_rtns_1 10000 463.8        46380.00 Numeric Functions/second
num_rtns_1 10000 463.9        46390.00 Numeric Functions/second

trig_rtns 10010 28.6713 286713.29 Trigonometric Functions/second
trig_rtns 10020 28.3433 283433.13 Trigonometric Functions/second
trig_rtns 10000 28.4 284000.00 Trigonometric Functions/second

matrix_rtns 10000 5964.3       596430.00 Point Transformations/second
matrix_rtns 10000 5906.8       590680.00 Point Transformations/second
matrix_rtns 10000 5906.6       590660.00 Point Transformations/second

array_rtns 10010 13.6863          273.73 Linear Systems Solved/second
array_rtns 10010 13.2867          265.73 Linear Systems Solved/second
array_rtns 10050 13.5323          270.65 Linear Systems Solved/second

string_rtns 10060 9.04573          904.57 String Manipulations/second
string_rtns 10050 8.95522          895.52 String Manipulations/second
string_rtns 10060 8.94632          894.63 String Manipulations/second

mem_rtns_1 10000 27.7 831000.00 Dynamic Memory Operations/second
mem_rtns_1 10010 24.975 749250.75 Dynamic Memory Operations/second
mem_rtns_1 10030 24.327 729810.57 Dynamic Memory Operations/second

mem_rtns_2 10000 1632.5 163250.00 Block Memory Operations/second
mem_rtns_2 10000 1631.8 163180.00 Block Memory Operations/second
mem_rtns_2 10000 1618 161800.00 Block Memory Operations/second

sort_rtns_1 10020 33.6327 336.33 Sort Operations/second
sort_rtns_1 10010 33.0669 330.67 Sort Operations/second
sort_rtns_1 10000 33.1 331.00 Sort Operations/second

misc_rtns_1 10000 782.2 7822.00 Auxiliary Loops/second
misc_rtns_1 10000 738 7380.00 Auxiliary Loops/second
misc_rtns_1 10000 732.1 7321.00 Auxiliary Loops/second

dir_rtns_1 10000 85.8       858000.00 Directory Operations/second
dir_rtns_1 10000 96.9       969000.00 Directory Operations/second
dir_rtns_1 10000 96.9       969000.00 Directory Operations/second

shell_rtns_1 10020 25.9481           25.95 Shell Scripts/second
shell_rtns_1 10030 24.1276           24.13 Shell Scripts/second
shell_rtns_1 10030 23.9282           23.93 Shell Scripts/second

shell_rtns_2 10010 26.0739 26.07 Shell Scripts/second
shell_rtns_2 10000 24.1 24.10 Shell Scripts/second
shell_rtns_2 10030 24.0279 24.03 Shell Scripts/second

shell_rtns_3 10010 26.0739 26.07 Shell Scripts/second
shell_rtns_3 10000 24.1 24.10 Shell Scripts/second
shell_rtns_3 10030 24.0279 24.03 Shell Scripts/second

series_1 10000 31924.9      3192490.00 Series Evaluations/second
series_1 10000 31651.5      3165150.00 Series Evaluations/second
series_1 10000 31649.9      3164990.00 Series Evaluations/second

shared_memory 10000 2227.4       222740.00 Shared Memory Operations/second
shared_memory 10000 1987.5       198750.00 Shared Memory Operations/second
shared_memory 10000 1994.7       199470.00 Shared Memory Operations/second

^^^^ 2.5.5* is still slower then 2.4.19

tcp_test 10000 661.7        59553.00 TCP/IP Messages/second
tcp_test 10000 558.7        50283.00 TCP/IP Messages/second
tcp_test 10000 541.8        48762.00 TCP/IP Messages/second

^^^^ 2.5.5* is still slower then 2.4.19

udp_test 10000 1182.7       118270.00 UDP/IP DataGrams/second
udp_test 10000 972.3        97230.00 UDP/IP DataGrams/second
udp_test 10000 988.1        98810.00 UDP/IP DataGrams/second

^^^^ 2.5.5* is still slower then 2.4.19

fifo_test 10000 1207 120700.00 FIFO Messages/second
fifo_test 10000 1052.3 105230.00 FIFO Messages/second
fifo_test 10000 1083.1 108310.00 FIFO Messages/second
^^^^ 2.5.5* is still slower then 2.4.19

stream_pipe 10000 2418.6       241860.00 Stream Pipe Messages/second
stream_pipe 10000 2281.5       228150.00 Stream Pipe Messages/second
stream_pipe 10000 2258.8       225880.00 Stream Pipe Messages/second

^^^^ 2.5.5* is still slower then 2.4.19

dgram_pipe 10000 2357.8       235780.00 DataGram Pipe Messages/second
dgram_pipe 10000 2112.2       211220.00 DataGram Pipe Messages/second
dgram_pipe 10000 2039.6       203960.00 DataGram Pipe Messages/second

^^^^ 2.5.5* is still slower then 2.4.19

pipe_cpy 10000 3918 391800.00 Pipe Messages/second
pipe_cpy 10000 3139 313900.00 Pipe Messages/second
pipe_cpy 10000 3139.3 313930.00 Pipe Messages/second
^^^^ 2.5.5* is still slower then 2.4.19

ram_copy 10000 19338.7 483854274.00 Memory to Memory Copy/second
ram_copy 10000 19160.9 479405718.00 Memory to Memory Copy/second
ram_copy 10000 19155 479258100.00 Memory to Memory Copy/second

Signature · 5 lines

______________________________________________
http://www.linuxmail.org/
Now with POP3/IMAP access for only US$19.95/yr

Powered by Outblaze

Hans Reiser

2002-12-16 23:01 UTC #1 lore raw

Andrew and Chris, are these changes in performance definitely due to VM changes (and not some difference I am not thinking of between 2.5 and 2.4 reiserfs code)?

Hans

Quoted 270 lines · Paolo Ciarrocchi

Hi all,
hw is an Omnibook 6000 (laptop)
fs is Reiserfs
format:
2.4.19
2.5.51
2.5.52
add_double 10030 23.9282       430707.88 Thousand Double Precision Additions/second
add_double 10030 23.7288       427118.64 Thousand Double Precision Additions/second
add_double 10030 23.7288       427118.64 Thousand Double Precision Additions/second
add_float 10000 35.9 430800.00 Thousand Single Precision Additions/second
add_float 10010 35.5644 426773.23 Thousand Single Precision Additions/second
add_float 10010 35.5644 426773.23 Thousand Single Precision Additions/second
add_long 10010 22.1778      1330669.33 Thousand Long Integer Additions/second
add_long 10020 21.9561      1317365.27 Thousand Long Integer Additions/second
add_long 10020 21.9561      1317365.27 Thousand Long Integer Additions/second
add_int 10020 22.0559      1323353.29 Thousand Integer Additions/second
add_int 10020 21.8563      1311377.25 Thousand Integer Additions/second
add_int 10020 21.8563      1311377.25 Thousand Integer Additions/second
add_short 10000 55.4      1329600.00 Thousand Short Integer Additions/second
add_short 10000 54.9      1317600.00 Thousand Short Integer Additions/second
add_short 10000 54.9      1317600.00 Thousand Short Integer Additions/second
creat-clo 10040 19.4223 19422.31 File Creations and Closes/second
creat-clo 10010 87.2128 87212.79 File Creations and Closes/second
creat-clo 10000 75.3 75300.00 File Creations and Closes/second
^^^Here 2.5.* is _a lot_ faster than 2.4.19 but 2.5.52 is slower then 2.5.51
page_test 10000 123.9 210630.00 System Allocations & Pages/second
page_test 10000 105.7 179690.00 System Allocations & Pages/second
page_test 10010 105.095 178661.34 System Allocations & Pages/second
^^^^ 2.5.5* is still slower then 2.4.19
brk_test 10010 48.951 832167.83 System Memory Allocations/second
brk_test 10020 44.6108 758383.23 System Memory Allocations/second
brk_test 10010 43.8561 745554.45 System Memory Allocations/second
^^^^ 2.5.5* is still slower then 2.4.19
jmp_test 10000 4313.7      4313700.00 Non-local gotos/second
jmp_test 10000 4273.2      4273200.00 Non-local gotos/second
jmp_test 10000 4273.1      4273100.00 Non-local gotos/second
signal_test 10000 166.1       166100.00 Signal Traps/second
signal_test 10000 157.4       157400.00 Signal Traps/second
signal_test 10000 159.6       159600.00 Signal Traps/second
^^^^ 2.5.5* is still slower then 2.4.19
exec_test 10000 13.8 69.00 Program Loads/second
exec_test 10020 12.9741 64.87 Program Loads/second
exec_test 10030 12.9611 64.81 Program Loads/second
^^^^ 2.5.5* is still slower then 2.4.19
fork_test 10000 44.8 4480.00 Task Creations/second
fork_test 10020 23.9521 2395.21 Task Creations/second
fork_test 10020 29.0419 2904.19 Task Creations/second
^^^^ 2.5.5* is still slower then 2.4.19
link_test 10000 155.3 9783.90 Link/Unlink Pairs/second
link_test 10000 147.7 9305.10 Link/Unlink Pairs/second
link_test 10010 140.959 8880.42 Link/Unlink Pairs/second
^^^^ 2.5.5* is still slower then 2.4.19
disk_rr 10050 6.96517        35661.69 Random Disk Reads (K)/second
disk_rr 10110 7.51731        38488.63 Random Disk Reads (K)/second
disk_rr 10050 7.46269        38208.96 Random Disk Reads (K)/second
^^^^ 2.5.5* is faster then 2.4.19
disk_rw 10060 5.666 29009.94 Random Disk Writes (K)/second
disk_rw 10000 6.8 34816.00 Random Disk Writes (K)/second
disk_rw 10090 6.73935 34505.45 Random Disk Writes (K)/second
^^^^ 2.5.5* is faster then 2.4.19
disk_rd 10010 38.1618 195388.61 Sequential Disk Reads (K)/second
disk_rd 10020 38.024 194682.63 Sequential Disk Reads (K)/second
disk_rd 10010 38.0619 194877.12 Sequential Disk Reads (K)/second
disk_wrt 10100 8.51485 43596.04 Sequential Disk Writes (K)/second
disk_wrt 10070 9.43396 48301.89 Sequential Disk Writes (K)/second
disk_wrt 10080 9.3254 47746.03 Sequential Disk Writes (K)/second
^^^^ 2.5.5* is faster then 2.4.19
disk_cp 10120 7.31225        37438.74 Disk Copies (K)/second
disk_cp 10010 7.69231        39384.62 Disk Copies (K)/second
disk_cp 10090 7.63132        39072.35 Disk Copies (K)/second
^^^^ 2.5.5* is faster then 2.4.19
sync_disk_rw 16020 0.062422 159.80 Sync Random Disk Writes (K)/second
sync_disk_rw 18790 0.0532198 136.24 Sync Random Disk Writes (K)/second
sync_disk_rw 14570 0.0686342 175.70 Sync Random Disk Writes (K)/second
^^^^ 2.5.5* is faster then 2.4.19
sync_disk_wrt 11220 0.0891266          228.16 Sync Sequential Disk Writes (K)/second
sync_disk_wrt 10150 0.0985222          252.22 Sync Sequential Disk Writes (K)/second
sync_disk_wrt 10130 0.0987167          252.71 Sync Sequential Disk Writes (K)/second
^^^^ 2.5.5* is faster then 2.4.19
sync_disk_cp 10870 0.0919963 235.51 Sync Disk Copies (K)/second
sync_disk_cp 19580 0.102145 261.49 Sync Disk Copies (K)/second
sync_disk_cp 10050 0.0995025 254.73 Sync Disk Copies (K)/second
^^^^ 2.5.5* is faster then 2.4.19
disk_src 10000 118.2 8865.00 Directory Searches/second
disk_src 10010 110.889 8316.68 Directory Searches/second
disk_src 10010 123.477 9260.74 Directory Searches/second
^^^^ 2.5.5* is faster then 2.4.19 and 2.5.52 is faster then 2.5.51
div_double 10000 24.4 73200.00 Thousand Double Precision Divides/second
div_double 10020 24.1517 72455.09 Thousand Double Precision Divides/second
div_double 10010 24.1758 72527.47 Thousand Double Precision Divides/second
div_float 10010 24.3756        73126.87 Thousand Single Precision Divides/second
div_float 10010 24.1758        72527.47 Thousand Single Precision Divides/second
div_float 10010 24.1758        72527.47 Thousand Single Precision Divides/second
div_long 10020 19.9601        17964.07 Thousand Long Integer Divides/second
div_long 10020 19.7605        17784.43 Thousand Long Integer Divides/second
div_long 10010 19.7802        17802.20 Thousand Long Integer Divides/second
div_int 10020 19.9601        17964.07 Thousand Integer Divides/second
div_int 10010 19.7802        17802.20 Thousand Integer Divides/second
div_int 10020 19.7605        17784.43 Thousand Integer Divides/second
div_short 10030 19.9402        17946.16 Thousand Short Integer Divides/second
div_short 10020 19.7605        17784.43 Thousand Short Integer Divides/second
div_short 10010 19.7802        17802.20 Thousand Short Integer Divides/second
fun_cal 10000 62.5     32000000.00 Function Calls (no arguments)/second
fun_cal 10000 61.9     31692800.00 Function Calls (no arguments)/second
fun_cal 10000 61.9     31692800.00 Function Calls (no arguments)/second
fun_cal1 10000 170.4 87244800.00 Function Calls (1 argument)/second
fun_cal1 10010 168.831 86441558.44 Function Calls (1 argument)/second
fun_cal1 10010 168.831 86441558.44 Function Calls (1 argument)/second
fun_cal2 10000 112.5 57600000.00 Function Calls (2 arguments)/second
fun_cal2 10000 111.5 57088000.00 Function Calls (2 arguments)/second
fun_cal2 10010 111.389 57030969.03 Function Calls (2 arguments)/second
fun_cal15 10010 34.0659     17441758.24 Function Calls (15 arguments)/second
fun_cal15 10010 33.7662     17288311.69 Function Calls (15 arguments)/second
fun_cal15 10010 33.7662     17288311.69 Function Calls (15 arguments)/second
sieve 10450 0.861244            4.31 Integer Sieves/second
sieve 10600 0.849057            4.25 Integer Sieves/second
sieve 10600 0.849057            4.25 Integer Sieves/second
mul_double 10020 21.5569 258682.63 Thousand Double Precision Multiplies/second
mul_double 10020 21.3573 256287.43 Thousand Double Precision Multiplies/second
mul_double 10030 21.336 256031.90 Thousand Double Precision Multiplies/second
mul_float 10030 21.5354 258424.73 Thousand Single Precision Multiplies/second
mul_float 10030 21.336 256031.90 Thousand Single Precision Multiplies/second
mul_float 10020 21.3573 256287.43 Thousand Single Precision Multiplies/second
mul_long 10000 947.7       227448.00 Thousand Long Integer Multiplies/second
mul_long 10000 939.6       225504.00 Thousand Long Integer Multiplies/second
mul_long 10000 939.8       225552.00 Thousand Long Integer Multiplies/second
mul_int 10000 951.9       228456.00 Thousand Integer Multiplies/second
mul_int 10000 943.2       226368.00 Thousand Integer Multiplies/second
mul_int 10000 943.3       226392.00 Thousand Integer Multiplies/second
mul_short 10000 759.1       227730.00 Thousand Short Integer Multiplies/second
mul_short 10000 753.4       226020.00 Thousand Short Integer Multiplies/second
mul_short 10000 753.4       226020.00 Thousand Short Integer Multiplies/second
num_rtns_1 10000 467.2        46720.00 Numeric Functions/second
num_rtns_1 10000 463.8        46380.00 Numeric Functions/second
num_rtns_1 10000 463.9        46390.00 Numeric Functions/second
trig_rtns 10010 28.6713 286713.29 Trigonometric Functions/second
trig_rtns 10020 28.3433 283433.13 Trigonometric Functions/second
trig_rtns 10000 28.4 284000.00 Trigonometric Functions/second
matrix_rtns 10000 5964.3       596430.00 Point Transformations/second
matrix_rtns 10000 5906.8       590680.00 Point Transformations/second
matrix_rtns 10000 5906.6       590660.00 Point Transformations/second
array_rtns 10010 13.6863          273.73 Linear Systems Solved/second
array_rtns 10010 13.2867          265.73 Linear Systems Solved/second
array_rtns 10050 13.5323          270.65 Linear Systems Solved/second
string_rtns 10060 9.04573          904.57 String Manipulations/second
string_rtns 10050 8.95522          895.52 String Manipulations/second
string_rtns 10060 8.94632          894.63 String Manipulations/second
mem_rtns_1 10000 27.7 831000.00 Dynamic Memory Operations/second
mem_rtns_1 10010 24.975 749250.75 Dynamic Memory Operations/second
mem_rtns_1 10030 24.327 729810.57 Dynamic Memory Operations/second
mem_rtns_2 10000 1632.5 163250.00 Block Memory Operations/second
mem_rtns_2 10000 1631.8 163180.00 Block Memory Operations/second
mem_rtns_2 10000 1618 161800.00 Block Memory Operations/second
sort_rtns_1 10020 33.6327 336.33 Sort Operations/second
sort_rtns_1 10010 33.0669 330.67 Sort Operations/second
sort_rtns_1 10000 33.1 331.00 Sort Operations/second
misc_rtns_1 10000 782.2 7822.00 Auxiliary Loops/second
misc_rtns_1 10000 738 7380.00 Auxiliary Loops/second
misc_rtns_1 10000 732.1 7321.00 Auxiliary Loops/second
dir_rtns_1 10000 85.8       858000.00 Directory Operations/second
dir_rtns_1 10000 96.9       969000.00 Directory Operations/second
dir_rtns_1 10000 96.9       969000.00 Directory Operations/second
shell_rtns_1 10020 25.9481           25.95 Shell Scripts/second
shell_rtns_1 10030 24.1276           24.13 Shell Scripts/second
shell_rtns_1 10030 23.9282           23.93 Shell Scripts/second
shell_rtns_2 10010 26.0739 26.07 Shell Scripts/second
shell_rtns_2 10000 24.1 24.10 Shell Scripts/second
shell_rtns_2 10030 24.0279 24.03 Shell Scripts/second
shell_rtns_3 10010 26.0739 26.07 Shell Scripts/second
shell_rtns_3 10000 24.1 24.10 Shell Scripts/second
shell_rtns_3 10030 24.0279 24.03 Shell Scripts/second
series_1 10000 31924.9      3192490.00 Series Evaluations/second
series_1 10000 31651.5      3165150.00 Series Evaluations/second
series_1 10000 31649.9      3164990.00 Series Evaluations/second
shared_memory 10000 2227.4       222740.00 Shared Memory Operations/second
shared_memory 10000 1987.5       198750.00 Shared Memory Operations/second
shared_memory 10000 1994.7       199470.00 Shared Memory Operations/second
^^^^ 2.5.5* is still slower then 2.4.19
tcp_test 10000 661.7        59553.00 TCP/IP Messages/second
tcp_test 10000 558.7        50283.00 TCP/IP Messages/second
tcp_test 10000 541.8        48762.00 TCP/IP Messages/second
^^^^ 2.5.5* is still slower then 2.4.19
udp_test 10000 1182.7       118270.00 UDP/IP DataGrams/second
udp_test 10000 972.3        97230.00 UDP/IP DataGrams/second
udp_test 10000 988.1        98810.00 UDP/IP DataGrams/second
^^^^ 2.5.5* is still slower then 2.4.19
fifo_test 10000 1207 120700.00 FIFO Messages/second
fifo_test 10000 1052.3 105230.00 FIFO Messages/second
fifo_test 10000 1083.1 108310.00 FIFO Messages/second
^^^^ 2.5.5* is still slower then 2.4.19
stream_pipe 10000 2418.6       241860.00 Stream Pipe Messages/second
stream_pipe 10000 2281.5       228150.00 Stream Pipe Messages/second
stream_pipe 10000 2258.8       225880.00 Stream Pipe Messages/second
^^^^ 2.5.5* is still slower then 2.4.19
dgram_pipe 10000 2357.8       235780.00 DataGram Pipe Messages/second
dgram_pipe 10000 2112.2       211220.00 DataGram Pipe Messages/second
dgram_pipe 10000 2039.6       203960.00 DataGram Pipe Messages/second
^^^^ 2.5.5* is still slower then 2.4.19
pipe_cpy 10000 3918 391800.00 Pipe Messages/second
pipe_cpy 10000 3139 313900.00 Pipe Messages/second
pipe_cpy 10000 3139.3 313930.00 Pipe Messages/second
^^^^ 2.5.5* is still slower then 2.4.19
ram_copy 10000 19338.7 483854274.00 Memory to Memory Copy/second
ram_copy 10000 19160.9 479405718.00 Memory to Memory Copy/second
ram_copy 10000 19155 479258100.00 Memory to Memory Copy/second

Andrew Morton

2002-12-16 23:18 UTC #2 lore raw

Quoted 5 lines · Hans Reiser

Andrew and Chris, are these changes in performance definitely due to VM changes (and not some difference I am not thinking of between 2.5 and 2.4 reiserfs code)?

aim9 is just doing

for (lots)
close(creat(filename))

ie: deleting a zero-length file.

So I changed inode_setattr() to not call into the filesystem at all if i_size was not being altered. Not a very interesting optimisation, but it was simple to do.

It seems that reiserfs_truncate() has a lot of setup overhead.

Andrew Morton

2002-12-16 23:30 UTC #3 lore raw

Quoted 8 lines · Andrew Morton

Quoted 5 lines · Hans Reiser
Andrew and Chris, are these changes in performance definitely due to VM changes (and not some difference I am not thinking of between 2.5 and 2.4 reiserfs code)?

Quoted 4 lines

aim9 is just doing
for (lots)
close(creat(filename))

unlink(filename); /* of course */

Hans Reiser

2002-12-16 23:52 UTC #4 lore raw

Quoted 12 lines · Andrew Morton

Quoted 9 lines · Andrew Morton
Quoted 6 lines · Hans Reiser
Andrew and Chris, are these changes in performance definitely due to VM changes (and not some difference I am not thinking of between 2.5 and 2.4 reiserfs code)?

Quoted 11 lines

Quoted 6 lines
aim9 is just doing
for (lots)
close(creat(filename))
unlink(filename); /* of course */

Oh, commercial fs vendors must really love tuning for this benchmark.... sigh....

Hans

Cliff White

2002-12-18 18:39 UTC #5 lore raw

Quoted 20 lines

Quoted 18 lines · Andrew Morton
Quoted 15 lines · Andrew Morton
Quoted 6 lines · Hans Reiser
Andrew and Chris, are these changes in performance definitely due to VM changes (and not some difference I am not thinking of between 2.5 and 2.4 reiserfs code)?
aim9 is just doing
for (lots)
close(creat(filename))

Quoted 8 lines

Quoted 5 lines
unlink(filename); /* of course */
Oh, commercial fs vendors must really love tuning for this benchmark.... sigh....

Ya, we think the AIM stuff is getting a little old. The basic idea is fine, but many of the tests do _very little work. We (OSDL) would like to re-do AIM9+7 and make it more useful. We'd love to have some input from everyone.... For example, how big a file should we create for a real creat() test ? cliffw

Quoted 8 lines

Hans
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Nikita Danilov

2002-12-18 19:02 UTC #6 lore raw

Quoted 25 lines · Cliff White

Quoted 25 lines
Quoted 23 lines · Andrew Morton
Quoted 9 lines · Andrew Morton
Quoted 6 lines · Hans Reiser
Andrew and Chris, are these changes in performance definitely due to VM changes (and not some difference I am not thinking of between 2.5 and 2.4 reiserfs code)?
Quoted 6 lines
aim9 is just doing
for (lots)
close(creat(filename))
unlink(filename); /* of course */

Quoted 7 lines

Quoted 3 lines
Oh, commercial fs vendors must really love tuning for this benchmark.... sigh....
Ya, we think the AIM stuff is getting a little old. The basic idea is fine, but many of the tests do _very little work. We (OSDL) would like to re-do AIM9+7 and make it more useful. We'd love to have some input from everyone.... For example, how big a file should we create for a real creat() test ?

Probably it should be variable. You may take a look at the way "mongo" script (http://www.namesys.com/benchmarks/mongo.tar.gz) used to benchmark reiserfs calculates size of file (reiser_fract_tree.c:determine_size() function). Hmm, I remember putting long comment for this function, but now it is gone...

Quoted 4 lines

cliffw
Quoted 2 lines
Hans

Nikita.

Hans Reiser

2002-12-23 07:05 UTC #7 lore raw

In reply to Cliff White

Quoted 37 lines · Cliff White

Quoted 37 lines
Quoted 33 lines · Andrew Morton
Quoted 13 lines · Andrew Morton
Quoted 8 lines · Hans Reiser
Andrew and Chris, are these changes in performance definitely due to VM changes (and not some difference I am not thinking of between 2.5 and 2.4 reiserfs code)?
Quoted 8 lines
aim9 is just doing
for (lots)
close(creat(filename))
unlink(filename); /* of course */

Quoted 13 lines

Quoted 4 lines
Oh, commercial fs vendors must really love tuning for this benchmark.... sigh....
Ya, we think the AIM stuff is getting a little old. The basic idea is fine, but
many of the tests do _very little work. We (OSDL) would like to re-do AIM9+7 and make it more useful. We'd love to have some input from everyone.... For example, how big a file should we create for a real creat() test ?
cliffw

Well, if you take a look at mongo.pl available at www.namesys.com we provide you with a fractal file size generation program that you might want to look at, that mongo uses during the creation portion of its benchmark.

It makes 80% of the files less than some amount (100 bytes, or 1k, or 4k, depending on how mean you want to be to ext2;-) ), then 80% of the remaining 20% less than 10 times that amount then....

There is more than one version of the file size generation code, so make sure you got the one that works as I described above. I am looking for a formula to smooth the above behavior into some sort of gentle curve rather than sharp bands, but I haven't found it yet.

Hans