The tag has no usage guidance.

learn more… | top users | synonyms

0
votes
0answers
13 views

Not able to understand the output of FFSB (Filesystem benchmark) ON RAID-6

I'm trying to run FFSB benchmark (https://sourceforge.net/projects/ffsb/) a RAID-6 server and the output is really confusing to evaluate. Distribution - Red Hat Linux kernel version - 2.6.32-504.30....
0
votes
0answers
31 views

running iperf over ssh

I'm trying to benchmark the network connection between 2 ubuntu servers on which I don't have root privileges. I did manage to download and start iperf-2.0.9 on both boxes. My problem is I can't ...
0
votes
1answer
14 views

Benchmark shared folder on Virtualbox?

I'm trying to benchmark what I can read and write from the shared folder on my Virtualbox machine is? From my local host machine I have a shared folder, and on the guest virtualized machine, I would ...
1
vote
0answers
18 views

Write-benchmark of gnome-disks

Does the write-benchmark of gnome-disks write back the original data after the test? It is destructive? The Benchmark Settings window only states this: Please back up important data before using the ...
1
vote
2answers
45 views

How to benchmark different shells [closed]

I want to know if shell 1 is better for a job than shell 2 with a test that I handle maybe with shell 3 for example ksh, dash, bash or zsh. I use a test script that builds my projects, starts it,feed ...
2
votes
0answers
34 views

Benchmarking under SELinux Sandbox with Phoronix Test Suite

I would like to run the Phoronix Test Suite 6.2.2 under the SELinux sandbox on Fedora 23, but I can't execute it using $ sandbox ./phoronix-test-suite. I probably don't have the correct SELinux type. ...
0
votes
0answers
29 views

Multicore benchmark

I'm looking for a way to benchmark a multi-core CPU. I found this #!/bin/bash echo "scale=$1; a(1)*4" | bc -l The first argument is number of digits to calculate. Now we can run in parallel (4 ...
1
vote
0answers
68 views

What do the benchmarks in HardInfo mean?

The program hardinfo is included by default with Lubuntu as a system profiler. It can run 6 different types of benchmarks: -CPU Blowfish -CPU Cryptohash -CPU Fibonacci -CPU N-Queens -FPU FFT -...
0
votes
0answers
37 views

Random write on DISK benchmark shows high read rate on iostat

I am running a random write benchmark directly on some disks (HDD & SDD) and I am getting some weird output data on the iostat monitor. Besides the usual write rate, there is a read rate that I ...
2
votes
0answers
16 views

confused about dd write speed and VFS page cache

After reading some articles on the Linux VFS page cache and the tunable parameters like dirty_ratio i was under the impression that page cache would operate as both read and write caching layer. But ...
1
vote
1answer
338 views

howto interpret `cryptsetup benchmark` results?

summary: I can run cryptsetup benchmark and sort the results, but seek guidance in their interpretation. E.g., should I give more weight to encryption speed or decryption speed? Should key-derivation ...
0
votes
0answers
29 views

What are useful options for testing a usb drive with fio?

I would test some usb drives using fio. I started with the tutorial on linux.com, but it was mainly for hard disk drives. This is the test I use for my hard drives: [random-read] rw=randread size=...
3
votes
1answer
57 views

Benchmarking With SELinux

I've been performing some benchmarking tasks on freshly installed CentOS machines with SELinux enabled and disabled. The tasks include copying files of various sizes, sending variable amounts of data ...
1
vote
0answers
24 views

How can I benchmark or audit an embedded kernel?

Is there a reliable or industry standard way to benchmark or audit embedded kernel performance? The target of the audit is to determine the time that the kernel spends in certain critical section, ...
0
votes
0answers
106 views

HPL benchmark show very low calculation speed

I’m constructing cluster for mph parallel calculation. To check performance of my node, I used HPL linpack benchmark. However, the benchmark result is much more low than a theoretical value. The ...
0
votes
0answers
34 views

IO benchmark tools block size impact?

Many IO benchmark tools (fio, dd, etc...) provide way to set block size. Lets say that I change 4K to 64k and test read/write on mounted SATA disk. I will get higher data throughput in sequential ...
1
vote
1answer
20 views

Where to find information about tests from Phoronix test suite?

Phoronix Test Suite is a linux benchmarking tool made of tests and suites of tests. To print all available test phoronix-test-suite list-available-tests To print all available suites phoronix-...
2
votes
0answers
34 views

GNOME Disks Sample Size

I'm trying to use GNOME Disks to test my disk performance and, according to my previous experience, I'd like to set the sample size to 8K, 16K and 32K. However, I see the minimal sample size in its ...
2
votes
0answers
12 views

How to force a dependency conflict for benchmarks?

I want to measure the time portage needs for solving dependency conflicts on a gentoo system. For a benchmark I need a reproducible conflict situation. It should have many solvable conflicts. /usr/...
0
votes
1answer
156 views

How can I benchmark NFS and SMB transfer speeds?

I'd like to compare the speeds of NFS and SMB across my network. Is there any good (and simple) tool to let me do this? Ideally I'd like something I can just point at my mounted filesystem (i.e. /...
1
vote
0answers
48 views

How can I profile/benchmark RAM performance?

I would like to know if my memory hardware is a bottleneck of my system performance. I am dealing with really old hardware (late 2000). If there is room for improvement, I would like to upgrade. ...
4
votes
2answers
622 views

Telling Linux kernel *not* to use certain CPUs

I'm trying to run some benchmarks on a multicore machine and I'd like to tell the Linux kernel to simply avoid certain cores unless explicitly told to use them. The idea is that I could set aside a ...
4
votes
2answers
260 views

What is actually the speed reported by dd?

I always thought that when doing: dd if=/dev/sdx of=/dev/sdy the report which is displayed once the command finishes, such as: 79304319+0 records in 79304319+0 records out 40603811328 bytes (41 GB)...
0
votes
2answers
650 views

How to benchmark memory in an embedded linux device?

Is there a generic utility to benchmark external memory device (eMMC, DDR) throughput on an embedded device (e.g. raspberry pi)? memtester, bonnie, bonnie++, memtest, dd? I've been running into ...
3
votes
0answers
154 views

How to generate graphs from lmbench results?

I need to use the lmbench benchmarking tool to get the system performance statistics. I am able to execute and run the results. In the results folder this is a MakeFile with this information # ...
0
votes
3answers
135 views

Which commands or operations can be used to put the cpu under intense load?

I would like to measure latencies during high cpu usage but I don't have benchmarking tools like stress and I can't install them. Which operations do you recommend to put the cpu under intense load? ...
2
votes
1answer
605 views

Simple floating point benchmark that can be compiled from source?

I am interested in benchmarking some ARM platforms; can anyone suggest a simple benchmark with source code that can be used to benchmark floating point performance? A package such as a .deb file ...
5
votes
4answers
453 views

How to measure Firefox's startup time?

I would like to measure how much time it takes for some softwares to start. I know the command time but in the case of firefox (iceweasel for me on Debian), this command prints only the time I let ...
1
vote
1answer
133 views

question about installing Rally on OpenStack

I have a quick question that I'm going to install Rally to benchmark OpenStack and I've read it's documents here. Now, I don't know if I install Rally on a vm in OpenStack that there's no service on ...
1
vote
1answer
210 views

Can I use fio on a mounted device?

I'm using fio to get some broad read IOPS performance data for various storage configurations like this: fio --name=readiops --filename=/dev/md1 --direct=1 --rw=randread --bs=4k --numjobs=4 --iodepth=...
3
votes
1answer
85 views

Using SLURM without a feature

Suppose my super computer has the following NODELIST's with the included features: NODELIST FEATURES NodeA (none) NodeB specialfeature and I am trying to benchmark ...
20
votes
3answers
2k views

Why is coreutils sort slower than Python?

I wrote the following script to test the speed of Python's sort functionality: from sys import stdin, stdout lines = list(stdin) lines.sort() stdout.writelines(lines) I then compared this to the ...
1
vote
0answers
61 views

How fast is a process reading/writing virtual memory?

We have top, htop, iotop, iostat, vmstat, etc, to measure many different things on Linux, but how can one measure how much MB/s a process is reading and writing to/from RAM, i.e., virtual memory?
1
vote
0answers
38 views

How can we measure performance overhead resulting from writing to files or folders?

I have a tool which takes as input a folder or a file and then it logs operations done on that file or folder (modifying,deletion,creating new folders or files, etc). I was looking for a way to ...
0
votes
1answer
558 views

iozone runs fail with “No such file or directory” errors

I want to start using iozone to benchmark an aufs filesystem on a virtual machine on ubuntu. I tried following the manual to use relevant parameters, but I keep getting a weird error on most runs. I ...
1
vote
1answer
242 views

Measuring transfer speed of USB drive

I want to write a program to find out the average data transfer speed of my USB port. How do I go about implementing it? The only method I can think of is writing a large file to the path where the ...
8
votes
2answers
3k views

Simulate an unkillable process in D state

For catastrophe testing scenarios on out server environment we're looking for an easy way to make a process stuck in D (uninterruptible sleep) state. Any easy ways? An example C sample code would be ...
3
votes
2answers
868 views

Read all files in folder and subfolders - progress and size

Command I have is: time find . -type f -print -exec cp {} /dev/null \; This command finds all files in current folder and subfolders, print the name of each file and copy each of them to /dev/null. ...
1
vote
1answer
262 views

ext4 to btrfs conversion: how many times the whole volume will be read during the process?

I want to convert the 3TB 80% full ext4 block device into btrfs and I want to estimate the time it will take. I understand that this process is disk-bound rather then cpu- or memory-bound. So can ...
3
votes
1answer
3k views

RFC 2544 / Y1564 software implementation

Is there any RFC 2544 and/or Y1564 software available for Linux? I'm aware that tools like iperf, bw-ng and bmon exist, but I was not able to find something that mimics Ethernet tester functionality ...
1
vote
2answers
377 views

dd command: Finding read/write throughput of a flash/SD?

I am running petalinux from sd card. And trying to find data copying rate from Sd card and from QSPI flash that is there on the Xilinx Zynq board. As suggested here , I am doing like this: root@...
2
votes
2answers
510 views

Is there a good benchmark for virtual memory performance?

Aside from running memory benchmarks, has anyone used a benchmark that stresses the virtual memory subsystem? I'm curious to see how performance differs in virtual machines with and without EPT ...
25
votes
3answers
66k views

How can I benchmark my HDD?

I've seen commands to benchmark one's HDD such as this using dd: $ time sh -c "dd if=/dev/zero of=ddfile bs=8k count=250000 && sync" Are there better methods to do so than this?
7
votes
2answers
354 views

How can other processes affect measurements made with `time`?

If I time a process using the time command, I get output for 'real', 'user', and 'sys'. My understanding from this discussion is that 'real' is wall time, whereas 'user' and 'sys' are process time. ...
3
votes
3answers
1k views

Best way to benchmark different encryption solutions on my system

I want to compare different encryption solutions for encrypting my system, possibly different solutions for different parts of the system such as /usr or /home. In particular I look at dm-crypt ...
7
votes
2answers
8k views

Benchmark ssd on linux: How to measure the same things as crystaldiskmark does in windows

I want to benchmark a ssd (possibly with encrypted filesystems) and compare it to benchmarks done by crystaldiskmark on windows. So how can I measure approximately the same things as ...
0
votes
1answer
606 views

Floating point performance in Linux

I've tested Geekbench for my computer for both Windows and Linux. It seems that Linux loses markedly in floating point performance. Is there anything I can do about it? Install an optimization ...
3
votes
1answer
246 views

Bandwidth improvements expected from switching to Wayland or Mir?

I was intrigued by this answer regarding why X is not scalable. I'm unable to find any specific speed comparisons between x forwarding on these different protocols. How much % gain in network speeds ...
3
votes
1answer
2k views

Benchmarking Apache using ab

I was doing some performance testing for Apache using ab. Apache in our case is only working as a load balancer balancing load between two Liferay servers. Initially, I did the testing for 1000 ...
1
vote
2answers
208 views

how to uniformly execute benchmarks on linux

I am doing project on adapting Linux for scalable multi core architecture. I am testing Linux's performance with benchmarks. When I am running the benchmarks in the system then other programs like ...