Questions tagged [performance]
The speed at which a program runs
968
questions
0
votes
0
answers
21
views
2TB NVMe slow in Linux only after booting Windows
My SSD is a 2TB PCIe 3.0 x4 M.2 NVMe (Mega Fastro MS200, some Korean brand, on sale under $100 in Germany). According to Windows smi_nvme_flash_id.exe utility it is a SM2262EN controller with Intel ...
- 1
0
votes
0
answers
21
views
Single core slow on i9 vs i3
Upgraded CPU from Intel Core i3-8350K to i9-9900K.
Intel Core i3-8350K @ 4.00 GHz 1 Processor, 4 Cores, 4 Threads.
Base clock rate 4.00GHz
GenuineIntel Family 6 Model 158 Stepping 11
Intel Core i9-...
- 13
1
vote
0
answers
16
views
Optimize recovery after memory pressure
I have a machine that compiles large codebases, with 32 threads and 16 GiB RAM.
The RAM is on the small side, but usually sufficient. When it isn't, however, the machine swaps out a few GiB of memory, ...
- 4,344
-1
votes
1
answer
23
views
http latency wild variance on localhost
Running on MBP M1 with 10 cores.
Running 1 thread client (wrk) and 1 thread server (java).
A youtube is also running on the system.
Load average 2 out of 10.
Except GC, which could be the reasons of ...
- 99
1
vote
0
answers
24
views
How is the hardware TX queue selected by default?
NICs may have multiple hardware RX and TX queues, and there are several methods which allow choosing which RX and TX queues are used for a given flow/packet, though most are focused on the RX side.
...
- 528
0
votes
0
answers
33
views
Is it possible to artificially trigger memory pressure on Linux?
I know that I can monitor / subscribe for events in case kernel is experiencing memory pressure using the memory pressure API. However, how about the opposite case? Is it possible to tell the Linux ...
- 3,779
0
votes
1
answer
55
views
I'm wondering why using cp in this way takes over 4 seconds [duplicate]
time cp /usr/local/bin/* /media/username/flash-drive-label
real 0m4.142s
user 0m0.007s
sys 0m0.007s
Normally using cp with wildcard (even when there are hundreds of files) like that never ...
- 167
1
vote
0
answers
79
views
Is there any reason to use NFS 3 over version 4.2?
Consider a work environment with 100gbps type Infiniband switches, running on enterprise class servers having 512 GB of RAM and larger, with the operating system being RHEL 8.7 or very close to it.
...
- 5,205
2
votes
2
answers
101
views
Why does higher CPU-usage slow down a task?
I am using whisper.cpp to transcribe some sound files. It is a very CPU heavy process so I try to find some optimal settings and therefore I have done some tests with the thread setting (-t) but the ...
- 1,591
3
votes
2
answers
343
views
Is SSD performance hindered by access to folder on SSD via symlink on conventional spinning rust HD?
I prefer to keep my OS and applications on one partition and /home/ on another.
I intend to replace the current / drive with an SSD, then symlink /home/myuser/.local/ to a directory on the SSD because ...
- 15.2k
1
vote
1
answer
34
views
DPROBES (DTRACE_PROBE) for measuring high latency stuff under 1µsec
Currently, I'm analyzing the performance of a high latency application but I'm not confident in my measurements at all. So far, I have used DPROBES for instrumentation and BCC/funclatency for ...
- 113
0
votes
1
answer
32
views
How to diagnose a sharp, sudden increase in the frequency with which processes enter the D ("uninterruptible sleep") state
The Linux cluster I work on recently started getting frozen for a few minutes at a time. I've determined that the reason for this behavior is that processes are very frequently landing in the D (&...
- 14.2k
1
vote
0
answers
47
views
host hangs after cat /dev/null > bigfile.log
I found a big log file(2.7 TB) on my disk, so I decided to empty it with the following command:
cat /dev/null > bigfile.log
After I executed this cmd, I lost my ssh connection. When I logged in ...
- 11
0
votes
0
answers
25
views
Limit write speed on disk based on remaining free space (on linux)
Is there a way to limit the overall write speed on a disk based on its remaining free space?
For example, if it is 99% full limit the speed such as filling the remaining space would require an entire ...
luca
0
votes
0
answers
92
views
WizTree equivalent for ext2/3/4
There are many applications for visualizing the files/directories that take up the most space, for example all of these: How to understand what's taking up disk space?, and other similar ones for ...
- 397
0
votes
1
answer
23
views
Does cygserver improve Cygwin performance? If so, with what tasks?
I've read that cygserver can improve performance in some circumstances but I'm not really clear on how - or on how to know whether it's applicable to our use case.
Can anyone provide any insights?
- 221
1
vote
0
answers
29
views
Clean ways to prime the kernel inode cache after boot?
Various answers on various StackExchange sites cover parts of this topic, but I did not see an exact match for this scenario. How can you identify every mounted standard filesystem, and traverse all ...
- 11
0
votes
0
answers
239
views
Why is DM-Integrity so slow compared to BTRFS?
I want to detect silent corruption of block devices similar to how BTRFS does that for files. I'd even like to do that below BTRFS (and disable BTRFS's native checksumming) so that I can tweak more ...
- 260
0
votes
0
answers
85
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 ...
0
votes
0
answers
26
views
Is resuming from hibernation I/O bound or CPU bound?
I use hibernation (with encryption) on my laptop, which has 16 GB of RAM, and a fast CPU (Ryzen 6700U).
When resuming, the system takes a while to load the image (13/14 seconds).
Is resuming from (...
- 859
0
votes
1
answer
131
views
504 Gateway Time Out on Load Testing on Apache Instance
I have a Laravel app deployed over Apache Instance
Instance Config is as follows,
T3A.2xLarge (vCPU = 4, Memory 16 GIB)
I have increased apache timeout to 600 Seconds, configured mpm_prefork as ...
- 103
1
vote
0
answers
62
views
How to measure memory bandwidth of an app?
My coworkers are trying to figure out a performance problem. I suspect they're reading lots of unnecessary memory and hitting a memory bandwidth problem but I don't want to suggest that unless I know ...
- 11
0
votes
0
answers
11
views
Getting neat results with perf
I use perf with the following commands:
perf record -o perf.data --call-graph dwarf executable
perf report -g
But the output I get is not very good.
I see instead this output:
How can I get such ...
- 241
2
votes
0
answers
118
views
Inode size 512 and 1024 bytes functions in ext4 and its pros and cons? (Need a official reference)
When i first trying and installing openSUSE tumbleweed, i see when formatting a ext4 i can choose the inode size 128, 256, 512, or 1024 bytes. Typically it is 256. I am thinking a larger inode size ...
0
votes
0
answers
49
views
How should I properly measure running time of a program?
I'm trying to measure performance of two implementations of a program (e.g progv1 and progv2 that do the same thing) using the time command, however not only they differ, but if I run time progv1 10 ...
- 479
0
votes
1
answer
38
views
List of applications using dsync or is dsync used in production?
We want to use dsync in production server, Is it production ready? I want to know the list applications using dsync. can someone provide me the list of applications which are using dsync.
Source code: ...
- 544
0
votes
2
answers
37
views
Is there an equivalent of procs_running for threads?
For right sizing Linux hosts, it's important to know the backlog of things that are pending. procs_running (particularly when divided by the number of cores) is a good indicator of process backlog, ...
- 962
1
vote
1
answer
60
views
How can I efficiently read entries in a directory?
I would like my program to efficiently detect if a file was added to a directory between two invocations. That analysis might block user input (this would be for the fish shell), so I want it to run ...
1
vote
0
answers
21
views
Need to measure pairwire (ip based) bandwidth used over time when using TC
I need to measure the datarates of packets between multiple servers. I need pairwise bandwidths between the servers (if possible even the ports), not the overall datarate per interface on each server.
...
1
vote
0
answers
36
views
napi/eth%d-8193 takes > 30% CPU
I have recently upgraded my kernel. Now napi/eth%d-8193 takes > 30% CPU:
top - 16:29:57 up 9 days, 4:16, 70 users, load average: 5.48, 4.84, 5.00
Tasks: 1935 total, 2 running, 1906 sleeping, ...
- 32.7k
1
vote
0
answers
126
views
How to tune Linux kernel to try to keep free RAM ready?
I have a system with very spikey memory usage and lots of disk activity. As such, I'd like to have files in the cache as much as possible but I'd like to have enough free RAM ready to go to quickly ...
- 3,779
1
vote
0
answers
156
views
Laptop CPU stuck at 800Mhz randomly but works again when unplugged
I have a Dell Precision 7550 with an Intel Xeon W-10885M (8 cores, 16 threads) that has been displaying a strange behavior lately. I have a dual boot between Windows 10 and Ubuntu 20.04.4 LTS with ...
- 171
1
vote
1
answer
193
views
How to change the default BDI `max_ratio` and/or `min_ratio` for all devices?
I'm using a system with multiple storage devices with different write throughput. As explained in question Why were "USB-stick stall" problems reported in 2013? Why wasn't this problem ...
- 3,779
0
votes
2
answers
48
views
Is it possible to chain two consecutive find commands?
I want to:
find all directories that start with a capital letter
then, find all directories named '.git' in those directories
This is for performance optimization. It's my solution to reduce the ...
- 499
0
votes
0
answers
265
views
Lower NVME performance on Debian
I have a NVME Reletech P400 SSD 512GB installed in my machine and I found out that under Linux (Debian) it performs significantly worse than under Windows.
I used Cross Platform Disk Test. Here it's ...
- 37
0
votes
1
answer
128
views
Debian 11 - How can i lower the load on my server?
I'm currently running a Debian 11 with a load average of 4.
I want to lower this load - But i don't really know where to start, like a first step on at least the analisis part...
It's a Docker Server, ...
- 163
3
votes
0
answers
234
views
EdgeRouter Lite OpenBSD traffic routing limits
When routing traffic between (virtual) interfaces on Ubiquiti EdgeRouter Lite it is hitting some sort of limit. I'd like to get some help to determine what sort of limit is that and how it can be ...
- 53
1
vote
1
answer
682
views
Seemingly random performance drops on Pop!_OS
I am sure many PC users have experienced slow downs on their computers over time (mine is about a year old), but this scenario I find strange.
The Problem
At about 5:30PM yesterday I was browsing the ...
- 121
1
vote
1
answer
473
views
Is increasing the max_pool_percent parameter for zswap a smart idea to combine concepts of zram and zswap to increase memory and decrease disk usage?
I don't know if I'm missing something, but please take this with a grain of salt.
So, zram is used to reduce I/O operations on the disk by making a swap on the memory which eliminates system freezes ...
- 11
1
vote
0
answers
85
views
How do diagnose occasional shell lag?
I am using zsh with a bunch of plugins (Python venv, git status, etc).
Occasionally, when I cd to a directory, it waits for a couple of seconds before letting me type in commands. When I cd to the ...
- 321
2
votes
0
answers
99
views
Question about transfer data between specific physical memory and NVME
I'm trying to transfer data from specific reserved physical memory to my NVME, IIUC, there are 3 steps:
In the kernel driver, the reserved-memory (specified physical address and size in dts) is ...
- 327
1
vote
1
answer
99
views
How to measure how long linux takes to shutdown
I need to calculate how long linux takes to shutdown. I am running an embedded busybox linux on ARM i.MX7. I looked in /var/volatile/log/messages but that only shows the last startup. I need to ...
- 274
0
votes
1
answer
588
views
How to debug Linux TCP slows/packet loss
I'm trying to track down some particular network paths which are slowing down to about 200KByte/sec. I see this performance through various tests including with scp, rsync and iperf3:
$ iperf3 -c 157....
- 143
0
votes
1
answer
141
views
WiFi an order of magnitude slower than in Windows using ath9k
I'm using an AR9485WB-EG wireless card with Ubuntu 22.04 and the transfer speeds are atrocious — everything from 0.3 to 12 Mbps, compared to 30 to 50 Mbps on Windows 7 running on the same hardware and ...
- 389
1
vote
1
answer
57
views
The most reliable way of testing XAMPP / Apache httpd performance [closed]
tl;dr What is the best / the most reliable way of testing and comparing performance of XAMPP / Apache server running on different computers and cloud services? If I need a high-level result (just a ...
- 733
1
vote
0
answers
155
views
Archlinux, 'kernel.perf_event_paranoid' it not respected
When reading the Performance Counters (PMC) on CPUS, the kernel.perf_event_paranoid must be <=1 (see Kernel doc)
The program below reads the PMC and should exit early with 1, if it cannot open the ...
- 141
-2
votes
1
answer
368
views
Why does the Linux kernel not use rust?
I have heard that the Linux Kernel uses rust. I don't know much about the programming language except that it has the potential to run faster than C.
According to websites Rust is more safe than C ...
0
votes
0
answers
176
views
Ubuntu18.04 - scaling_cur_freq is lower than scaling_min_freq
While using the server, we have confirmed that the clock speed is very low only for a specific CPU group. Could you please tell me the cause and solution for this?
All cpu's cpuinfo_min_freq and ...
- 1
1
vote
0
answers
63
views
How to determine cause of poor CPU performance - AMD 75F3?
I have 2 identical machines with pretty identical configurations, however one machine is performing much worse than the other. After some digging it appears the CPUs on one machine is simply slower ...
- 23
2
votes
1
answer
504
views
Why 7z not using enough CPU?
CentOS 9. 3 vCPU VM with 4Gb RAM.
I run a cron job with 7z compressing 35Gb data in 150 files 7za a -mx=9 -mmt=3 ...
RAM usage - 18%, disk queue is very small, CPU is 61% on average. Why not 100% ? ...
- 139