[erlang-questions] Matching IP address in socket module

Micael Karlberg micael.karlberg@REDACTED
Tue Oct 29 11:12:19 CET 2019


Hi,

Its not a secret :) You will find it in the emulator test catalog.
The only problem is that it was never intended for public "consumption"
and is therefor not documented:

	socket_test_ttest_tcp_socket.erl (uses socket.erl)
	socket_test_ttest_tcp_gen.erl    (uses gen_tcp.erl)

The idea here is that they should have an "identical" interface (so
the users should work with both transports).
Obviously the "socket" is different (a gen_tcp socket is a port and
a socket socket is a term()).

The other socket_test_ttest_tcp* modules "can" be used to see how to
use these...

Also, there is work on making gen_tcp work with the new socket module.
This is ongoing. There is a lot of special options that gen_tcp/inet
manages (packaging for instance).

Regards,
	/BMK

On 2019-10-29 10:54, Loïc Hoguin wrote:
> Hello,
> 
> I would like to do a similar experiment with RabbitMQ. Would you mind sharing the code that makes 
> 'socket' work similar to 'gen_tcp'? That may give me a head start, even if the code is not complete.
> 
> Thanks,
> 
> On 25/10/2019 12:39, Micael Karlberg wrote:
>> Hi,
>>
>> Its early days still, but the goal is definitely that it should
>> be faster.
>>
>> Here is some figures from a (time) test tool which is part of the
>> test suite (its basically a ping-pong case), both server and client
>> running on the same host (but in different VMs):
>>
>> Two transports:
>>          gen:  gen_tcp
>>          sock: socket (tcp)
>>
>> The socket active mode is a simulation of gen_tcp active mode
>> (no active = N).
>>
>> The tables are the exchange from the client side.
>>
>> With server side using gen (active = false):
>>
>>          Transport       Active          Data
>>          gen             false           10192 byte/ms, 154 msgs/ms
>>          gen             true            10383 byte/ms, 157 msgs/ms
>>          gen             once            6003 byte/ms,  90  msgs/ms
>>          sock            false           14050 byte/ms, 212 msgs/ms
>>          sock            true            14772 byte/ms, 223 msgs/ms
>>          sock            once            14050 byte/ms, 210 msgs/ms
>>
>>
>> With server side using gen (active = true):
>>
>>          Transport       Active          Data
>>          gen             false           9447 byte/ms,  143 msgs/ms
>>          gen             true            22345 byte/ms, 338 msgs/ms
>>          gen             once            5532 byte/ms,  83  msgs/ms
>>          sock            false           15316 byte/ms, 232 msgs/ms
>>          sock            true            23693 byte/ms, 358 msgs/ms
>>          sock            once            22068 byte/ms, 334 msgs/ms
>>
>>
>> With server side using sock (active = false, async = true):
>>
>>          Transport       Active          Data
>>          gen             false           11260 byte/ms, 170 msgs/ms
>>          gen             true            22273 byte/ms, 337 msgs/ms
>>          gen             once            7703 byte/ms,  116 msgs/ms
>>          sock            false           15211 byte/ms, 230 msgs/ms
>>          sock            true            24778 byte/ms, 375 msgs/ms
>>          sock            once            23086 byte/ms, 349 msgs/ms
>>
>>
>> With server side using sock (active = true, async = true):
>>
>>          Transport       Active          Data
>>          gen             false           11351 byte/ms, 171 msgs/ms
>>          gen             true            22469 byte/ms, 340 msgs/ms
>>          gen             once            7407 byte/ms,  112 msgs/ms
>>          sock            false           15484 byte/ms, 234 msgs/ms
>>          sock            true            24885 byte/ms, 377 msgs/ms
>>          sock            once            23570 byte/ms, 357 msgs/ms
>>
>>
>> There is of course a bit of overhead since the socket transport
>> is trying to emulate (part of) gen_tcp.
>>
>> This is run on a Dell Precision T1700 running SLES 12 SP2 (not that
>> that effects the relative performance).
>> Run with a snapshot from the maint branch.
>>
>> This is obviously not a real use case, but it can be a guideline.
>> Also no UDP test (at the moment).
>>
>> /BMK
>>
>> On 2019-10-24 18:48, Frank Muller wrote:
>>> Hi Thomas
>>>
>>> Is the socket module faster than gen_tcp/gen_udp in your case? If yes, can you please share some 
>>> numbers.
>>>
>>> /Frank
>>>
>>>      > case socket:recvfrom(Sock, 0, nowait) of
>>>      >       {ok, {#{family := Domain,
>>>      >               port   := Port,
>>>      >               addr   := Addr}, Data}} ->
>>>      >       .
>>>
>>>     Hi Micael, Mark,
>>>
>>>     thanks for your replies. The snippet above helped me getting the record
>>>     matched.
>>>
>>>     I'm quite impressed with the socket module, it seems to be pretty
>>>     complete, at least for my application.
>>>
>>>     Thanks,
>>>     Thomas
>>>
> 



More information about the erlang-questions mailing list