I have got a Cobalt RaQ 4r, the firmware was upgraded and Fedora Core 4 was installed. As part of the upgrade I wanted to replace the factory CPU with one that is listed as a drop in replacement... Of course, after spending
the money I wanted to see if there was any advantage. So, I found this
benchmark software and thought I would post the results.
CPU/Speed: AMD K6-2 450Mhz
Ram: 512 PC100 DIMM
Motherboard: Cobalt RaQ 4r
Bus: 1 PCI, 0 ISA
Cache: 64KB (on processor)
Controller: IDE ATA33
Disk: Seagate 20 Gig x 2 (Mirror)
Load: 1 user
Kernel: Linux 2.6
Kernel ELF?:
pgms:
options =
Code:
BYTE UNIX Benchmarks (Version 3.11)
System -- Linux s3 2.6.16.RAQ_GEN-III #44 Sun May 7 22:55:15 CEST 2006 i586 i586 i386 GNU/Linux
Start Benchmark Run: Sun Jul 23 06:04:39 EDT 2006
1 interactive users.
Dhrystone 2 without register variables 796159.5 lps (10 secs, 6 samples)
Dhrystone 2 using register variables 791246.8 lps (10 secs, 6 samples)
Arithmetic Test (type = arithoh) 1921095.7 lps (10 secs, 6 samples)
Arithmetic Test (type = register) 126941.5 lps (10 secs, 6 samples)
Arithmetic Test (type = short) 126041.4 lps (10 secs, 6 samples)
Arithmetic Test (type = int) 108394.8 lps (10 secs, 6 samples)
Arithmetic Test (type = long) 127342.3 lps (10 secs, 6 samples)
Arithmetic Test (type = float) 72524.3 lps (10 secs, 6 samples)
Arithmetic Test (type = double) 73025.5 lps (10 secs, 6 samples)
System Call Overhead Test 300940.4 lps (10 secs, 6 samples)
Pipe Throughput Test 145349.7 lps (10 secs, 6 samples)
Pipe-based Context Switching Test 39199.7 lps (10 secs, 6 samples)
Process Creation Test 779.7 lps (10 secs, 6 samples)
Execl Throughput Test 321.1 lps (9 secs, 6 samples)
File Read (10 seconds) 548164.0 KBps (10 secs, 6 samples)
File Write (10 seconds) 44181.0 KBps (10 secs, 6 samples)
File Copy (10 seconds) 12159.0 KBps (10 secs, 6 samples)
File Read (30 seconds) 545675.0 KBps (30 secs, 6 samples)
File Write (30 seconds) 45194.0 KBps (30 secs, 6 samples)
File Copy (30 seconds) 10564.0 KBps (30 secs, 6 samples)
C Compiler Test 116.9 lpm (60 secs, 3 samples)
Shell scripts (1 concurrent) 404.4 lpm (60 secs, 3 samples)
Shell scripts (2 concurrent) 251.0 lpm (60 secs, 3 samples)
Shell scripts (4 concurrent) 130.0 lpm (60 secs, 3 samples)
Shell scripts (8 concurrent) 66.0 lpm (60 secs, 3 samples)
Dc: sqrt(2) to 99 decimal places 11009.5 lpm (60 secs, 6 samples)
Recursion Test--Tower of Hanoi 15177.7 lps (10 secs, 6 samples)
INDEX VALUES
TEST BASELINE RESULT INDEX
Arithmetic Test (type = double) 2541.7 73025.5 28.7
Dhrystone 2 without register variables 22366.3 796159.5 35.6
Execl Throughput Test 16.5 321.1 19.5
File Copy (30 seconds) 179.0 10564.0 59.0
Pipe-based Context Switching Test 1318.5 39199.7 29.7
Shell scripts (8 concurrent) 4.0 66.0 16.5
=========
SUM of 6 items 189.0
AVERAGE 31.5
Now, from what I can tell... This Cobalt RaQ is the slowest of an of the computers in this thread!
Here are the results with the slick new processor:
CPU/Speed: AMD K6-III+ 450Mhz
Ram: 512 PC100 DIMM
Motherboard: Cobalt RaQ 4r
Bus: 1 PCI, 0 ISA
Cache: 256KB (on processor)
Controller: IDE ATA33
Disk: Seagate 20 Gig x 2 (Mirror)
Load: 1 user
Kernel: Linux 2.6.16
Kernel ELF?:
pgms:
options =
Code:
BYTE UNIX Benchmarks (Version 3.11)
System -- Linux s3 2.6.16.RAQ_GEN-III #44 Sun May 7 22:55:15 CEST 2006 i586 i586 i386 GNU/Linux
Start Benchmark Run: Sun Jul 23 08:12:58 EDT 2006
1 interactive users.
Dhrystone 2 without register variables 794883.8 lps (10 secs, 6 samples)
Dhrystone 2 using register variables 768329.3 lps (10 secs, 6 samples)
Arithmetic Test (type = arithoh) 2020875.3 lps (10 secs, 6 samples)
Arithmetic Test (type = register) 125523.5 lps (10 secs, 6 samples)
Arithmetic Test (type = short) 127377.5 lps (10 secs, 6 samples)
Arithmetic Test (type = int) 125572.7 lps (10 secs, 6 samples)
Arithmetic Test (type = long) 127743.5 lps (10 secs, 6 samples)
Arithmetic Test (type = float) 73081.9 lps (10 secs, 6 samples)
Arithmetic Test (type = double) 65772.6 lps (10 secs, 6 samples)
System Call Overhead Test 339040.1 lps (10 secs, 6 samples)
Pipe Throughput Test 213860.2 lps (10 secs, 6 samples)
Pipe-based Context Switching Test 65743.0 lps (10 secs, 6 samples)
Process Creation Test 1830.0 lps (10 secs, 6 samples)
Execl Throughput Test 570.4 lps (9 secs, 6 samples)
File Read (10 seconds) 585951.0 KBps (10 secs, 6 samples)
File Write (10 seconds) 52859.0 KBps (10 secs, 6 samples)
File Copy (10 seconds) 13910.0 KBps (10 secs, 6 samples)
File Read (30 seconds) 568939.0 KBps (30 secs, 6 samples)
File Write (30 seconds) 49347.0 KBps (30 secs, 6 samples)
File Copy (30 seconds) 13754.0 KBps (30 secs, 6 samples)
C Compiler Test 176.2 lpm (60 secs, 3 samples)
Shell scripts (1 concurrent) 611.3 lpm (60 secs, 3 samples)
Shell scripts (2 concurrent) 311.0 lpm (60 secs, 3 samples)
Shell scripts (4 concurrent) 167.7 lpm (60 secs, 3 samples)
Shell scripts (8 concurrent) 84.7 lpm (60 secs, 3 samples)
Dc: sqrt(2) to 99 decimal places 17305.3 lpm (60 secs, 6 samples)
Recursion Test--Tower of Hanoi 13789.7 lps (10 secs, 6 samples)
INDEX VALUES
TEST BASELINE RESULT INDEX
Arithmetic Test (type = double) 2541.7 65772.6 25.9
Dhrystone 2 without register variables 22366.3 794883.8 35.5
Execl Throughput Test 16.5 570.4 34.6
File Copy (30 seconds) 179.0 13754.0 76.8
Pipe-based Context Switching Test 1318.5 65743.0 49.9
Shell scripts (8 concurrent) 4.0 84.7 21.2
=========
SUM of 6 items 243.9
AVERAGE 40.6
If there are any people out there who are still using these old Cobalt servers (I know there are still lots in production), I also did some heat tests on the processor at the same time. With the CPU fan disabled (unplugged - but still using fin) at idle the stock processor was 35 C, running the benchmarks the temp went upto 65 which is still below the max temp of 85 for these K6 processors.
Doing the same tests, with no fan, the CPU at idle was about 40 During the benchmark it went to 68.5, so not much more. These measurements were taken with an infrared temperature monitor.
Hello everyone.
Does anyone know where to I could find published benchmarks for how a Linux box performs. It would be nice if I could find a comparison to the Windows OS.
Thanks,
Lance (2 Replies)
STEP 1: Get the source here:
https://www.unix.com/source/bm.zip
or
https://www.unix.com/source/unix_linux_bench.tar.gz
STEP 2: unzip or untar and cd into the bm directory
STEP 3: make
(Note: there is a pre-compiled Linux binary in the distro, so Linux users don't have to make a... (0 Replies)
I created two computers with identical hardware, and run the benchmark programs in both starting at the same exact time.
What makes no sense is that the computer that has the lower average index (121) finished the race a good 30 minutes ahead of the computer wich showed the higher avg index... (0 Replies)
My system bench mark results
INDEX VALUES
TEST BASELINE RESULT INDEX
Arithmetic Test (type = double) 2541.7 876123.7 344.7
Dhrystone 2 without register variables 22366.3 5411602.3 ... (1 Reply)
Hello,
I am using suse 10.1 and isql from unixodbc to connect to a MS SQL server. I got everything to work fine. What do I need to unload the results from the sql to a file? I attempted to put unload to in my sql statement but got a error. I don't see in the isql help where it has a option to... (0 Replies)
Here is my script:
#
# Capture the current directory.
export -p CurrentDir="`pwd`"
echo $CurrentDir
#
# Capture the new directory name in the form YYYYMMDD.
export -p DateDir="`date +"%Y%m%d"`"
echo $DateDir
#
# Store the desired target directory.
export -p... (6 Replies)
I seem to be somewhat baffled by the results of the -L option of the ls command on our new Linux system.
I have a symbolic link defined like the following:
ptmawpmfld.sqr as /u21/xxxx/m/moranp1/yyyy/fixpaths/tests20091218/awpmfld.sqr
When I issue the command ls -l ptmawpmfld.sqr, I see the... (2 Replies)
I am trying to sort a file . The file looks like this:
DDFF 2 /ztpfrepos/pgr/load
DDFQ 2 /ztpfrepos/pgr/load
DDFX 2 /ztpfrepos/pgr/load
DDUA 2 /ztpfrepos/pgr/load
My command:
sort -k1 /home/c153507/Bin/OPL1.txt -o /home/c153507/Bin/OPL1.txt
The results are OK except for one line where... (4 Replies)