From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tom Rymes To: development@lists.ipfire.org Subject: Re: Introducing fireperf Date: Thu, 04 Feb 2021 11:16:46 -0500 Message-ID: <04D30B6A-D429-4CD2-8881-69B53BEDA49D@rymes.net> In-Reply-To: <23B3120C-4854-4DAB-A3B8-41C44424834C@ipfire.org> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="===============5240384853745303750==" List-Id: --===============5240384853745303750== Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Michael, Any concerns that this could be used for evil? Tom > On Feb 4, 2021, at 10:57 AM, Michael Tremer w= rote: >=20 > Hello all y=E2=80=99all, >=20 > I would like to introduce a small side-project I have been working on calle= d fireperf. >=20 > It is a networking benchmarking tool which I have written to debug some iss= ues with IPFire on AWS and I thought this would be very useful for others, to= o. >=20 >=20 > A bit of a backstory >=20 > Everyone probably knows iperf and its newer brother iperf3 and has used the= m. They can do bandwidth tests and usually should be quite good at then. Unfo= rtunately they are reaching their limits quite early. In my environment I had= at least a 5 GBit/s connection between my two machines and I wanted to creat= e lots and lots of connections to stress-test the connection tracking. >=20 > This was unfortunately not possible with either of them, because iperf is s= tarting a thread per connection and iperf3 limits its connections to 128 per = process. In both cases this simply did not scale because my goal was to creat= e connections in the 6 figures or more range. I simply ran out of memory. >=20 > Another issue that both sometimes have - and which I did not validate speci= fically in this case - is that they cannot generate enough traffic to saturat= e a link. However, I need to be able to simply trust that this is possible as= long as I have the CPU resources available. >=20 > Therefore a new tool was needed. >=20 > When I started writing fireperf, I did not intend to make it fit for throug= hput tests, but since it was such a low-hanging fruit in the development proc= ess I added it, too. The original goal was simply to open a number of connect= ions (at least hundreds of thousands) and keep them open or let me know when = this is no longer possible. >=20 > Since I knew I was working on IPFire, I started to take advantage of Linux= =E2=80=99 modern APIs and try to delegate as much work as possible to the ker= nel. Especially since the whole Meltdown/Spectre debacle, sending data betwee= n the kernel and userland is slow, and the less work I have to do in the user= land, the more time I can spend on other things. >=20 > Therefore I use epoll() to let the kernel tell me when a socket is ready to= accept data and when something has happened and the connection broke down. I= am using getrandom() to get random data to send and I use timerfd to regular= ly notify me when to print some statistics. Therefore this application is not= very easily portable (it wasn=E2=80=99t an original design goal), but I am s= ure that there are some alternatives available if someone were to port this t= o another OS. >=20 > iperf3 - the most efficient one I knew - used up all of my 8GB of memory on= my test system when started multiple times to create about 10k connections. = Fireperf uses a few hundreds of kilobytes with tens of thousands of open conn= ections. In fact, it does not keep any state about the connections and theref= ore uses the same amount of memory no matter how many connections are open. T= he kernel will use some memory though, but I could not measure how much. >=20 > Without saturating my processor I can saturate any network link that I coul= d test up to 10 GBit/s. CPU usage normally is less than 10% and fireperf know= s a mode (=E2=80=94k) in which it won=E2=80=99t send any data, but only keep = the connections open and regularly let the kernel (again, because I am a lazy= developer) send some keep alive packets. That way, it uses next to no CPU re= sources while still generating a lot of stress for the network. >=20 > So here it is, my new tool. I hope someone finds this useful. >=20 > It is nice and tiny and everything comes in one binary file which only depe= nds on the C standard library. >=20 > Sources are available on our Git server as usual: >=20 > https://git.ipfire.org/?p=3Dfireperf.git;a=3Dsummary >=20 > I tagged release number 0.1.0 and I will push a patch into next very soon. = There are also Debian packages available if you want to give fireperf a try o= n Debian: >=20 > deb https://packages.ipfire.org/fireperf buster/ > deb-src https://packages.ipfire.org/fireperf buster/ >=20 > Replace buster with bullseye or sid if you are on those and do not forget t= o import the key: >=20 > curl https://packages.ipfire.org/79842AA7CDBA7AE3-pub.asc | apt-key add - >=20 > Documentation in form of a man-page is available here: >=20 > https://man-pages.ipfire.org/fireperf/fireperf.html >=20 > It would be great, if fireperf would become a great tool to benchmark IPFir= e. We definitely could do better in this department and hopefully gain better= insights on any regressions in performance, or if certain hardware is better= than other. I suppose throughput is not everything and fireperf should be ab= le to help us measure other factors, too. >=20 > -Michael --===============5240384853745303750==--