Skip to main content

Questions tagged [benchmark]

The tag has no usage guidance.

Filter by
Sorted by
Tagged with
22 votes
2 answers
1k views

Why are my benchmark times not repeatable, even for a CPU-bound task?

I'm running some benchmarks on my Linux desktop/laptop computer, but I'm not getting reliable results. I'm running a CPU-intensive task that does negligible I/O and doesn't use much RAM. My computer ...
Gilles 'SO- stop being evil''s user avatar
0 votes
0 answers
32 views

IOzone Filesystem Benchmark WEIRD Results on EXT4

I'm currently using an i7 with 32 GB of RAM to measure ext4 performance. However, I've encountered some unusual results when testing with 8 GB and 16 GB files. Shouldn't the performance increase ...
DemonioValero's user avatar
0 votes
1 answer
95 views

io_uring with `fio` fails on Rocky 9.3 w/kernel 5.14.0-362.18.1.el9_3.x86_64

I've tried various variations of the command: fio --name=test --ioengine=io_uring --iodepth=64 --rw=rw --bs=4k --direct=1 --size=2G --numjobs=24 --filename=/dev/sdc lower queue depth direct set to 1/...
Grant Curell's user avatar
0 votes
1 answer
64 views

Benchmarking lustre filesystem

I want to benchmark the ability of a single lustre client to save in its lustre-mounted filesystem. I am an application developer and not a storage maintainer, so i am not worried about storage write ...
Marco Montevechi Filho's user avatar
1 vote
1 answer
83 views

Why is the load of 'stress' unevenly distributed?

When I used stress to conduct stress testing on my system, I found that there was an issue with the distribution of CPU usage. CPU2 had a much higher usage rate than the other three CPUs. Excuse me, ...
Vimer's user avatar
  • 67
0 votes
1 answer
86 views

Which tools for micro-benchmarking?

I'm not sure which tool to use for micro-benchmarking a C program. I would like to measure both: Memory usage, RSS ( Resident Set Size ) CPU cycles I did use perf record -g and perf script piped ...
jinTgreater's user avatar
0 votes
0 answers
19 views

cset shield fails to execute task on a shielded core

I'm trying to run a benchmark on a shielded core on x86-64 (4 cores) and Ubuntu 22.04. Shielding core 3 runs fine: paul@cube-ubuntu-22:~/simd-agner/build$ sudo cset shield -c 3 cset: --> activating ...
Paul Jurczak's user avatar
0 votes
3 answers
81 views

Analogy of dd command real case when performing RW test

I'm doing these two tests: #1 dd if=/dev/zero of=./testfile1 bs=512M count=1 conv=fsync And #2 dd if=/dev/zero of=./testfile2 bs=1M count=512 conv=fsync Can I say: #1 is meaning copying (write ...
Muhammad Ikhwan Perwira's user avatar
1 vote
2 answers
167 views

Is there a performance penalty for backgrounding a process?

If I do: sleep 1 versus sleep 1 & wait $! will there be any difference in terms of CPU usage for spawning a foreground process versus a background process? Or will the performance of both lines ...
Maestro's user avatar
  • 191
0 votes
2 answers
104 views

How to figure out if CPU or GPU has been damaged by a thermal spike on Fedora

I have bought a new Asus Rog Strix laptop, and I ran on it Windows for a month without any issue. After I changed OS and putted Fedora 37 on it. My laptop has an NVidia 3060 with 6 GB (and a Ryzen 7 ...
Noumeno's user avatar
  • 183
0 votes
0 answers
560 views

understanding fio test results - 90th percentiles benchmark - Latency --> avg/stdev?

What are the relationships in storage benchmarking with fio between the average(avg) and standard deviation (stdev) at latency? Avg / stdev at latency is important for sequential or random test? Most ...
Hamza Karabulut's user avatar
0 votes
0 answers
109 views

How to check whether my network connection supports HTTP2, gRPC or WebSocket?

My ISP/AS/NAT is problematic since I think it blocks some transfer-layer protocols. I want to check whether my connection to my domain behind a CDN (e.g. cloudflare) supports HTTP/2, gRPC, QUIC or ...
Ragahito's user avatar
2 votes
0 answers
513 views

Is there an SSD Benchmark Tool for Linux to measure speed vs time or speed vs space usage?

Most of the SSD benchmarks you see out there use solely Crystal Disk Mark for Windows on an empty SSD. This allows to show speeds of 300~550MB/s (SATAIII), which is amazing compared to most HDDs. ...
Jeron Baffom's user avatar
-2 votes
2 answers
325 views

Using the command "sudo chrt -r 99" for a video encoding benchmark I always get bigger (worse) times

I am using a video encoding benchmarking application. The strange thing to me is that when I am using the command sudo chrt -r 99 VideoEncoding cfg I always get bigger time (worse results) instead of ...
Maverick's user avatar
  • 107
10 votes
1 answer
882 views

tee is faster than no tee

I have this simple bash script: #!/bin/bash for i in {1..1000000} do echo "hello ${i}" done That prints a message a million times. I try to compare the performance of dumping all the ...
areller's user avatar
  • 201
1 vote
1 answer
2k views

What exactly is being cached when opening/querying a SQLite database

I was asked to improve existing code to query SQLite databases. The original code made a lot of separate calls to the database and filtered the results in Python. Instead, I opted to re-write the ...
Bram Vanroy's user avatar
1 vote
1 answer
326 views

Is there a tool to count the latency of UDP datagrams in Linux?

Now I want to test the network performance between two Linux systems. The test scenario is as follows: one terminal sends UDP datagrams to the other terminal continuously at fixed intervals (the ...
fooh's user avatar
  • 11
0 votes
2 answers
340 views

Does the time command include the memory claimed by forked processes?

I want to benchmark some scripts with the time command. I am wondering if this command catches child processes' memory usage. command time -f '%M' python my_script.py If not, what are my options? Is ...
HappyFace's user avatar
  • 1,622
0 votes
0 answers
795 views

OSCollector Cannot run program "python": error=13, Permission denied

When I tend to install and use the tool named benchmarksql-5.0, I found this problem below [oracle@db01 run]$ ./runBenchmark.sh props.ora 09:43:33,081 [main] INFO jTPCC : Term-00, 09:43:33,083 [main]...
user17697264's user avatar
0 votes
0 answers
326 views

Which kernel functions are called upon page cache accesses?

I have an application and I'd like to measure its page cache hit/miss numbers. I found this tool called cachestat by Brendan Gregg, which uses the ftrace tool to count the number of 4 specific ...
WannabeArchitect's user avatar
1 vote
2 answers
205 views

Is there a tool that can read and write to disk at constant throughput?

I'm looking for a tool that can read and write data at a constant target throughput, say 3Mb/s, rather than pushing the I/O system to its limits. I then intend to monitor various metrics whilst this ...
Bruce Johnson's user avatar
1 vote
1 answer
4k views

stdint.h: no such file or directory

I am trying to compile coremarks to benchmark one of my CPU cores I generated (from here: https://gitlab.com/incoresemi/core-generators/benchmarks/-/tree/master). I get the following error: In file ...
thatbangaloreanguy's user avatar
1 vote
1 answer
62 views

How to purge in-memory disk cache

Linux uses free RAM to store pages of disk files to speed up overall disk access. However for the purposes of benchmarking, I want to test the disk read speed and I don't want that speed measurement ...
asker13's user avatar
  • 111
1 vote
0 answers
521 views

Is jq internal sort slower than GNU sort?

While filtering through this json file I did a benchmark and found out utilizing jq's internal sort and unique method is actually 25% slower than sort --unique! Command Mean [ms] Min [ms] Max [ms] ...
Zeta.Investigator's user avatar
1 vote
1 answer
1k views

testing network transfer speed and hardware limitations

This is somewhat of a computer engineering and network engineering question, but it is linux based since infiband is largely utilized via the linux operating system. And I am interested in ...
ron's user avatar
  • 6,944
1 vote
1 answer
1k views

help with iperf and infiniband and multiple NIC

I have two Dell servers in a rack in my server room: RHEL 7.9 x86-64 and iperf-2.0.13-1.el7.x86_64 hostnames are A and B each has 1 10GbE intel nic, having 2 ports 1 1Gbps intel nic {traditional} ...
ron's user avatar
  • 6,944
2 votes
0 answers
218 views

How to use kcbench?

I downloaded kcbench according to the online tutorial(https://gitlab.com/knurd42/kcbench). The result displayed after running the script is also correct. Processor: Intel(R) Xeon(R) CPU E5-...
user avatar
2 votes
1 answer
68 views

How to install kcbench on a computer without internet?

I want to install kcbench on the Linux server I use. The server is not connected to the Internet, but my computer (window) is able to connect to the Internet, and I can connect to the server via ssh. ...
user avatar
0 votes
1 answer
246 views

Are sysbench results comparable between versions and distros?

I have a machine which i can boot into either Ubuntu 18.04.5 or Fedora 32. I have had some performance issues, so i thought i would run a quick benchmark on both distros. I installed sysbench from the ...
Tom Anderson's user avatar
  • 1,008
3 votes
1 answer
761 views

Does it make sense to use queue-depth when doing synchronous IO benchmark?

Does it make sense to have a queue depth > 1 when doing a synchronous IO benchmark? I was expecting the same result a of QD1, but QD32 does give better result? I thought it would just be ignored. ...
MrCalvin's user avatar
  • 699
6 votes
1 answer
9k views

NVMe performance hit when using LUKS encryption

I'm running all the commands on the Proxmox host, when I LUKS encrypt a disk I get about a 28% hit in read speeds and 66% hit in write speeds. I have aes-ni enabled and I should be getting good enough ...
milkyway's user avatar
1 vote
0 answers
324 views

Cannot get sysbench to stop at max time

I am looking to stress test a machine using sysbench so that I can get the CPU temperature to write to a file each second. Right now the test runs past the max time that I set. How to get it to stop ...
Jinzu's user avatar
  • 109
-1 votes
1 answer
188 views

Forked process never terminates

I'm wanting to run sysbench and have the process fork in the background. When I run the code below. The stress test completes but I never can get back to bash prompt. Note, I only find the first prime ...
Jinzu's user avatar
  • 109
0 votes
1 answer
316 views

fio: how to write 2X user capacity?

My system has SSDs and I wanted to run some benchmark tests on it based on SSS PTS (SNIA). For example, for IOPS test, the spec suggested to use QD=32, TC=4 and for doing IO, the spec says following. ...
user3488903's user avatar
1 vote
1 answer
820 views

fio: how to reduce verbosity?

When I run fio command, I get huge file with following lines which fills up the entire space. I am interested in only the final fio output summary. How can I reduce this fio verbosity? :::: Jobs: 4 (f=...
user3488903's user avatar
9 votes
4 answers
15k views

Find the lightest desktop environment

A high-votes answer here for the question "What's the lightest desktop" which actually tried to quantitatively assess memory use relies on a Wikipedia page which quotes 2011 data. The newest ...
K7AAY's user avatar
  • 3,866
1 vote
3 answers
259 views

Where can I find a comparison of graphics cards using free software drivers? Are there any with good performance? [closed]

Is there a comparison of graphics cards that work with exclusively free software (FOSS drivers without any blobs)? If possible the comparison should include price and performance benchmarks of the ...
mYnDstrEAm's user avatar
  • 4,370
4 votes
1 answer
666 views

Measuring cumulative peak memory usage of multiple parallel processes

Imagine I run a bunch of processes in parallel using xargs and want to measure their cumulative peak memory usage. How can I do that in Linux? GNU time only measures peak RSS of a single process I run....
Ilia Minkin's user avatar
-1 votes
1 answer
427 views

How kernel building is a memory intensive application? [closed]

What makes benchmarks like kernel-build, pbzip, pfscan memory-intensive applications?
y_159's user avatar
  • 137
1 vote
2 answers
374 views

Why do Linux distributions have different performance?

Phoronix published a comparison of 9 operating systems. The three fastest are: Clear Linux 32310 CentOS Stream (while the rolling Manjaro is the penultimate) Fedora Workstation 31 What are basic ...
homocomputeris's user avatar
1 vote
1 answer
3k views

Is there anything similar to CrystalDiskMark for UNIX?

If you're shopping for an SSD, you've surely seen one of those screenshots from CrystalDiskMark with a few green squares, and the 2x4 matrix with results of doing read/write tests for the given ...
cnst's user avatar
  • 3,283
0 votes
1 answer
219 views

FIO benchmark replay fails in case of size 0

Abhishek Srivastava Member 0 1 post Report post Posted yesterday at 01:46 PM I have downloaded Financial application traces in standard *.spc format and then converted it into fio format. Now when ...
abhishek srivastava's user avatar
1 vote
1 answer
1k views

Does mixing up GTK and Qt applications slow down the machine?

Usually it is recommended to either use GTK or Qt apps in the operating system, but not mixing them both together. Example: When using KDE it seems to be obvious to use KMail (Qt) instead of ...
Dave's user avatar
  • 1,036
1 vote
1 answer
2k views

Printing milliseconds with GNU time

I am using GNU time for benchmarking and would like to measure real, user and sys time to the nearest millisecond. That is, I want to measure seconds to 3 decimal places, not the default 2. Does GNU ...
Jacob Baird's user avatar
2 votes
0 answers
57 views

Record filesystem activity

I would like to record all filesystem activity so that I can later replay it at will (for benchmarking purposes). What's the best way of doing that? I considered strace, but rejected that because it ...
Nikratio's user avatar
  • 121
3 votes
4 answers
6k views

Linux utility to bench mark clock speed of CPU

I have a Linux PC with a 3.4 GHz CPU. I must check this processor to see if it actually runs at that speed. Is there a benchmark available? I ran sysbench but it only provides time of completion, and ...
Hafiz Muhammad Shafiq's user avatar
2 votes
1 answer
4k views

FIO reporting slow write speeds while DD reports fast ones

According to most disk benchmarks, sequential write speeds are typically very close to the read speeds. Mine are in the ~500MB/s range. $ dd if=/dev/zero of=tempfile bs=1M count=5120 conv=notrunc ...
Cestarian's user avatar
  • 2,071
1 vote
1 answer
103 views

What does the CPU consumption in top mean when a CPU is assigned to a single process?

Being on Linux I have "CPUAffinity= 0 1 3" set in system.conf. I also use "taskset -cp 2 $pid" to assign a CPU to a single process. But what does the CPU consumption for the process in top now mean? ...
affincpu's user avatar
2 votes
1 answer
964 views

Why do diskspd and fio generate odd numbers on Linux vs. Windows?

UPDATE Thanks to Anon's answer, I figured out that the file system was at fault - I used NTFS. The following are the results using FAT32. Windows: diskspd64 -b128K -d5 -o32 -t1 -W0 -Sh -w0 cdm 508, ...
AndyO's user avatar
  • 173
1 vote
1 answer
2k views

How does FIO benchmark set IOdepth?

NOTE: My question stems from this other U&L Q - What exactly is iodepth in fio? I want to know how internally FIO sets I/O depth. I.e., one of the parameters we submit to FIO when we run it is "...
ahmadi's user avatar
  • 13