The performance tag has no wiki summary.
1
vote
2answers
7 views
Very poor read performance compared to write performance on md(raid1) / crypt(luks) / lvm
I'm experiencing very poor read performance over raid1/crypt/lvm. In the same time, write speeds are about 2x+ faster on the same setup. On another raid1 setup on the same machine I get normal read ...
0
votes
0answers
25 views
If a buggy application underperforms, it it possible server tuning could have no affect?
To set the stage, imagine if you will:
A read-heavy Tomcat Application with MySQL on Red Hat Enterprise Linux with some problematic Java Class files which resulted in the application performing a ...
1
vote
0answers
26 views
Monitoring Disk Latency
We have quite a few physical machines in our environment, yet we have no trending data over a period of time for disk latency statistics. We can easily run the standard iostat commands to see what the ...
2
votes
1answer
25 views
What is the expected performance of obnam? Or: why is it so slow?
I've been playing around with obnam these last few days, and although it looks very promising and seems to offer basically everything I ever wanted in a backup tool, I am pretty disappointed in its ...
1
vote
1answer
25 views
CPU Placement in KVM
I have a configuration of KVM guests where the cpu placement is configured as follows:
<vcpu placement='static'>1</vcpu>
Now since I don't have a cpuset specified I was reading libvirt ...
0
votes
1answer
34 views
Is it useful to setup RAID0 on laptop? [closed]
I have a Dell laptop, which I am going to replace the CD-ROM hub with a HDD. I am thinking if I setup a RAID0 on my original HDD and the new HDD, would I have some I/O performance improvement?
...
1
vote
1answer
36 views
How does downloading speed vary between download managers? [closed]
I used several download managers on Ubuntu and found variation in downloading speed.
uGet (recommended)
Flareget
xdm (Xtreme Download Manager)
Steadyflow
I find variation in downloading speed.
My ...
3
votes
1answer
61 views
Performance differences between pipelines and process substitution
I tend to use pipelines in my bash scripts over process substitution in most situations, especially in cases of using multiple sets of commands as it seems more readable to do ... | ... | ... over ... ...
2
votes
1answer
56 views
speed up linux server [closed]
On my VPS, I have centOS 6.5 installed with 2 virtual cpu 2048 MB memory, 60 GB disk and 300 GB monthly data. Doing a speed test today resulted in 168.62 Mbit/s dload and 37.39 Mbit/s upload.
Then my ...
1
vote
0answers
24 views
is star faster than GNU and BSD tar? [closed]
I made serveral runs (create a tarball) on OS X Mavericks with BSD tar, GNU tar and star. star is always the winner.
System is iMac (Mid 2010) only running Terminal.
root@time var> du -sh log/
...
3
votes
2answers
117 views
How to record the actual running time of a program with other programs running? [duplicate]
When timing the wall time performance of an OpenMP program on Linux while other programs are running, how can I get the actual running time?
2
votes
4answers
159 views
Why is writing so slow on raw device, and so fast on filesystem (USB key)?
I have a USB key that I am trying to benchmark quick'n'dirty on Linux. I'm testing write bandwith.
dd on raw partition
I created a partition starting at sector 2048, then did a 4 GB sequential ...
5
votes
1answer
100 views
Attempting to Abbreviate kSAR's PDF output
I'm currently writing a cronjob around kSar's command line (non-GUI) options in order generate PDF's. I'm just going to email a PDF of the previous day's activity to a different webmaster (basically ...
3
votes
1answer
121 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 ...
12
votes
1answer
1k views
Why is my TCP throughput much greater than UDP throughput?
I haven't done anything unusual to my hardware or kernel configurations (all default settings, fresh OS install, Linux kernel 3.11 TCP/IP stack) and I'm averaging about 3.83 million messages per ...
1
vote
1answer
180 views
High performance routing with high churn ruleset. What approach is most likely to scale to C10K and beyond?
I've been asked to create a high throughput router using PC / server hardware. I'm modestly familiar with pf, iptables, and tc, but I've never had such high churn and so many rules. I hope you can ...
23
votes
5answers
3k views
Why is iterating over a file twice faster than reading it into memory and computing twice?
I'm comparing the following
tail -n 1000000 stdout.log | grep -c '"success": true'
tail -n 1000000 stdout.log | grep -c '"success": false'
with the following
log=$(tail -n 1000000 stdout.log)
echo ...
1
vote
1answer
91 views
Can we make grep faster by indexing words in files? [duplicate]
Over a decade ago I imagined an improved version of grep that would use indexes to make recursive searching faster. I never got around to writing it, but I wonder if somebody else did!
My original ...
35
votes
1answer
2k views
Diagram of Linux kernel vs. performance tools?
Is there a diagram that shows how the various performance tools such as ip, netstat, perf, top, ps, etc. interact with the various subsystems within the Linux kernel?
6
votes
1answer
94 views
Do processes in cgroups have to be set every boot? If so, how to persist them?
I was reading this answer to a question on setting limits for processes by using cgroups. The sample given sets limits for a process (in the example, sshd). In the answer, the PID of the process is ...
6
votes
7answers
493 views
Make disk/disk copy slower
Is there a method of slowing down the copy process on Linux?
I have a big file, say 10GB, and I'd like to copy it to another directory, but I don't want to copy it with full speed. Let's say I'd like ...
2
votes
2answers
111 views
Why load is high despite the fact that neither CPU or disk is overused
I'm getting the following output from top:
Cpu(s): 43.8%us, 32.5%sy, 4.8%ni, 2.0%id, 15.6%wa, 0.2%hi, 1.2%si, 0.0%st
Mem: 16331504k total, 15759412k used, 572092k free, 4575980k buffers
...
2
votes
3answers
62 views
How to optimize this oneliner? (if second column empty, put the line in another file)
I have a little "oneliner":
while read ONELINE; do
if [[ -n "`echo $ONELINE | awk '{print $2}'`" ]]; then
echo "$ONELINE" >> LIST-ok.txt
else
echo "$ONELINE" >> ...
2
votes
1answer
113 views
Analyse the performance of an rsync command from FreeNAS to a USB drive
I infrequently run a task where I rsync my photos (~30 directories with 10-100 files in each) from my FreeNAS machine to an external USB hard drive (500GB WD) as a redundant backup.
I execute the ...
3
votes
3answers
228 views
Can I just disable updatedb?
Is updatedb necessary at all? I never use locate and my servers tend to have dozens of millions of files which usually makes updatedb to run for a long time and consume I/O needed by MySQL and/or ...
0
votes
1answer
398 views
What is the effect of rsync --block-size on large files?
I want to backup a few terabytes of large files (500mb to 1500mb each) to a backup server (BULK storage) via rsync.
For this I use the following command (updated with suggestions from @Gilles):
...
1
vote
0answers
14 views
Amount of memory transfer on a running Linux
Can I determinate the amount of RAM exchange (between the CPU and the RAM) on a running Linux system?
I'm not looking for the max RAM speed (then memtest would be valid answer). I'm looking for the ...
1
vote
1answer
56 views
How to measure the time a packet spends on the system(router)?
I have a streaming system and one if its components is a router/firewall. I'm using iptables for setting the required behaviour on the router.
I'm having performance issues in forms of a constant ...
4
votes
2answers
371 views
Why is dd using direct slower writing to disk than to a file
I am trying to compare aggregate write rates when writing to a file in a GPFS file system, as compared to writing directly to a disk on a system with Red Hat Enterprise Linux Server release 6.4 ...
1
vote
0answers
32 views
memory consumption alert
My VPS has reported a memory consumption alert says reached 273% and I cannot access any of my sites.
But I checked on the terminal today and it does not look like a memory problem. Am I correct in ...
3
votes
1answer
118 views
CPU is free, yet bash script not utilizing all CPU resources
I ran a simple script to generate a large (10000000 lines) csv file with 6 fields in which some fields changed in each line/row, using a while loop. The machine had all (32) CPUs free, lot of RAM (~31 ...
2
votes
1answer
136 views
What is the most accurate way to measure disk read/write speed for an OpenVZ VPS host from within the host itself?
CentOS 6.x | OpenVZ
I have an openvz VPS host and would like to measure disk read/write speed. Historically, on other physical systems, I've ran hdparm to collect this information. I haven't been ...
2
votes
1answer
91 views
Does negative vs positive matching impact firewall performance?
I am curious if using negative vs. positive matching impacts the performance of the netfilter stack. e.g. is iptables -I INPUT -p tcp -s 192.168.0.0/16 -j DROP equivalent to iptables -I INPUT -p tcp ...
0
votes
1answer
75 views
ERROR: cube viewer is not available!
I have installed and configured the following:
cube-4.2.1
opari2-1.1.1
papi-5.2.0
scalasca-2.0
scorep-1.2.1
I am trying to run scalasca -examine, but I get the following error:
ERROR: cube viewer ...
4
votes
2answers
446 views
Why is rm slow?
I have been trying to use rsnapshot for making backups, but I'm finding it unusable. While it is able to diff a directory (50gb) and duplicate it (hardlinking every file) in a few minutes, and I can ...
3
votes
2answers
360 views
Basic sed command on large one-line file: couldn't re-allocate memory
I have a 250 MB text file, all in one line.
In this file I want to replace a characters with b characters:
sed -e "s/a/b/g" < one-line-250-mb.txt
It fails with:
sed: couldn't re-allocate ...
7
votes
2answers
49 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.
...
1
vote
1answer
104 views
Performance of RAID 1 with 3 Disks under Linux
The standard way to setup a RAID 1 under Linux is to use two disks of the same size.
The write performance is almost as good as with a single disk drive.
The read performance can be up to twice the ...
1
vote
0answers
82 views
Solaris 10 on x86: local sockets (unix domain sockets) very very slow? [closed]
I'm writing an application both on Linux and Solaris 10. It uses local sockets for some internal communications. On Linux an open, exchange one short message & close cycle takes about 10us elapsed ...
0
votes
2answers
36 views
Fedora 17 Performance Issues [closed]
I have a Dell Latitude E4310. I have Windows 7 and Fedora installed on it. I rarely use windows and I have been using fedora for a good 9 months now. Recently the performance on my machine degraded ...
1
vote
1answer
738 views
Running thousands of curl background processes in parallel in bash script
I am running thounsand of curl background processes in parallel in the following bash script
START=$(date +%s)
for i in {1..100000}
do
curl -s "http://some_url_here/"$i > $i.txt&
...
3
votes
4answers
203 views
Does cron jobs per se represent a significant burden on a modern system?
I'd like to run a pretty simple script (is there a dir with name x in dir y if so move it to dir z–and x will only be there about once a day) every 15 seconds or so (one minute divided into 4*15 ...
2
votes
1answer
42 views
Multiple Lan interfaces on one VM rather than multiple VM
I'm looking if it's possible to give rights, to only allow the use of one network interface to a specific user on a Linux distribution.
My use case: I want, for my studies, testing the ...
1
vote
2answers
61 views
Varing speed when reading sequentially from a HDD
I have just got a new HDD and I wanted to test the sequential read speed (the whole disk). I used:
pv /dev/sda > /dev/null
The disk is 500GB WD VelociRaptor (WD5000HHTZ).
At the beginning the ...
0
votes
1answer
161 views
Performance Issue: AWS Medium Linux Instance V/S Hetzner Dedicated Instance
I am testing an AWS Medium Instance with CentOS (AWS AMI) and comparing it with Hetzner's EX40-SSD (http://www.hetzner.de/hosting/produkte_rootserver/ex40ssd). Software setup on both is comparable( ...
23
votes
9answers
875 views
for vs find in Bash
When looping through files there are two ways:
1. use a for-loop
for f in *; do
echo "$f"
done
2. use find
find * -prune | while read f; do
echo "$f"
done
Assuming these two loops will ...
1
vote
1answer
152 views
Ubuntu 12.04LTS becoms slow and jerky randomly for a mintue and then becomes normal by itself [closed]
I am using Ubuntu 12.04LTS on my Toshiba satellite L510 and I have only Ubuntu installed. The system becomes slow and jerky by itself after every 10minutes or so and the remains like that for less ...
9
votes
2answers
659 views
Why is wc so slow?
Why is the wc utility so slow?
When I run it on a large file, it takes about 20 times longer than md5sum:
MyDesktop:/tmp$ dd if=/dev/zero bs=1024k count=1024 of=/tmp/bigfile
1024+0 records in
1024+0 ...
3
votes
3answers
162 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 ...
3
votes
1answer
204 views
Efficient way to print lines from a massive file using awk, sed, or something else? [duplicate]
If I had a plain-text file containing 8 million lines and I want to print out lines 4,000,010 to 4,000,000 to the screen, which would be more efficient: awk or sed?
There is no pattern to the text, ...
