35
In benchmarking you must always establish what your limits are. Because if you expect to get 100mbs out of that lan than you are only fooling your self!
Look at this Block design of the RaspberryPI Model-B
So we establish a very important fact here. Ethernet is bottlenecked by the USB controller because form the block digram we establish it is connected to ...
28
I expect that, as Alex says, the benchmarks will show that the fastest Linux webservers will still be the fastest, regardless of architecture.
If anyone wants to run benchmarks then the following tutorial has been useful to me:
How to perform benchmarks on a web-server
Serving Static Pages
I have tested the RPi using Apache serving a simple static page:
...
24
You can perform a Raspberry PI stresstest with the sysbench tool.
First example calculate primes
sysbench --test=cpu --cpu-max-prime=20000 run
Second example test the I/O Output of your Raspberry Pi
sysbench --test=fileio --file-total-size=2G prepare
sysbench --test=fileio --file-total-size=2G --file-test-mode=rndrw --init-rng=on --max-time=300 --max-...
9
I wrote the little command-line tool stressberry the other day which stresses your Raspberry, measures the core temperature, and produces nice plots. Install with
pip install stressberry --user
and run with
stressberry-run out.dat
stressberry-plot out.dat -o out.png
6
For the CPU and most other things you can use
nice -19 stress-ng -c 4 --metrics --timeout 60s
For the GPU they suggest GeeXLab at geeks3d.com.
6
Whbn the foundation announced the enhanced overclocking support they mentioned running quake 3 was a good test of the improved performance. This forum post has the installation details.
5
Apache is not the best choice when serving static content, nginx is better suited for that. I did a benchmark using http://lekensteyn.nl/index.html as test document. It is a larger document than Jiving's example since 19 bytes is not realistic for an actual page. The results are quite amazing, the RPi even outperforms my work laptop (maybe because I have all ...
4
There are some benchmarks from Henner Zeller's repository on GitHub which claimed that directly outputting data to the GPIO could achieve up to 65.8 MHz on a Raspberry Pi 3 (not B+, mind, but I suspect the figures won't be that far off). The code used is available here in C and the author gives the following pseudocode equivalent:
// Pseudocode
for (;;) {
...
4
Running that command to bench your RPi is not so acurate. That command will burn 1/4 cores of your Pi, thats why you get weirds results. Try to activate all 4 cores at 100%.
I coded a small benchmark script for raspberry (multi-core benchmark), to test it just copy and paste this command on your RPi:
curl https://raw.githubusercontent.com/aikoncwd/rpi-...
4
Iperf is an excellent program for measuring maximum bandwidth between networks.
#Sever listening on TCP port 1337
ip route
iperf -s -p 1337
#Sever listening on UDP port 1337
iperf -s -u -p 1337
#Client connecting to server IP(TCP)
iperf -c <IP_ADDR> -p 1337 -t <TIME> -i <PING_TIME> -f m -d
#Client connecting to server IP(UDP)
iperf -c <...
4
XBMC doing multimedia decoding might be a good test as well as the UI there is in OpenGL. A game that constantly hammers the frame buffer (memory!) while including a numerical simulation like OpenTTD is also a pretty good test. The generic Dhrystone and Whetstone benchmarks might also be good. There's an OpenGL game I'm working on porting called Armagetron ...
3
Obviously, the highest performance would be with a ramdisk.
That won't be much use for a network-attached-storage device, but then, the pi is not necessarily the best choice for that fixed application.
For some things where the pi's flexibility is uniquely suited, actual storage requirements may be small enough to run explicitly from a ramdisk, or perhaps ...
3
The SD card interface is unchanged.
See Raspberry Pi 2 - FAQ and collated answers
3
There are several factors to consider, not just card speed.
As one person as stated, the Raspberry Pi could be the bottle neck, yet that could be its hardware, or software driver. Be advised, I'm using my Raspberry Pi ONLY to learn Mathematica; my advice on this comes from an electronics background and extensive IT support over 3 decades.
I also looked at ...
3
Well structured question. Try using:
hdparm -t /dev/sdb
Assuming your device is on /dev/sdb. Also, perhaps your devices just have equal read speeds? Do you have any others to test out?
2
To measure the sequential read or write speed, you can use dd.
Write speed:
dd if=/dev/zero of=test.tst bs=4096 count=100000
Read speed:
dd if=test.tst of=/dev/null bs=4096 count=100000
then remove the test file:
rm test.tst
Measuring is tricky, writing small files will get a benefit from caching in RAM, and when reading the data may be in RAM already ...
2
It's very likely that the RPi itself it bottlenecking the SD card, making it irrelevant what the speed of the card is, if it can't interface it fast enough.
2
If you want to synchronously transmit data at 50MHz frequency (that's how I read "pins to be toggled on the schedule of a ~50 mhz FPGA clock"), you will absolutely need to use the same clock source for the FPGA and whatever device you connect to it. RPi doesn't have programmable clock inputs/outputs, so it's absolutely not up to the task.
Your best bet is ...
2
Does Raspberry Pi Model B+ have sufficient CPU and memory to handle both mdadm and encrypted hdds?
I'm not an expert on, or even a user of such things, but unless this places significant load on your desktop machine (e.g, runs one core at 25%+ all the time), then sure.
The pi has a single core, 700 Mhz processor. I don't think the I/O can happen fast ...
2
For what it's worth, I just ran that same test on a stock Raspberry Pi 3B running Raspbian Jessie and it completed in 24.3 seconds. It took 73.4 s on a Model B rev 2 running Wheezy.
Unless you have an idea of the system load beforehand, this isn't a particularly good general benchmark. Arbitrary-precision decimal number handling is a slightly special case. ...
1
I have recently been through this process and wanted to capture the answer for future reference. These benchmarks were taken using a regular stopwatch.
Configure: 30 minutes
Make: 80 minutes
Make Install: 1 minutes
I was building CMake version: 3.13
using the following steps:
git clone https://github.com/Kitware/CMake.git
cd CMake
mkdir build &&...
1
I expect this can be made to 'work' (you will need 4 x USB-SATA 'dongles'), however throughput will be hopelessly slow ...
All 4 USB ports & Etrhernet (on the B+ or a A/B with external Hub) are linked to a single port on the GPU and all i/o is handled by the CPU byte at a time = unlike PC which has 'stand-alone' interface chips that support DMA direct ...
Only top voted, non community-wiki answers of a minimum length are eligible
