[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: [tor-bugs] #16614 [metrics-lib]: Improve pipelining during data-read, data-parse, parsed-write.

#16614: Improve pipelining during data-read, data-parse, parsed-write.
-----------------------------+---------------------
     Reporter:  leeroy       |      Owner:  karsten
         Type:  enhancement  |     Status:  new
     Priority:  normal       |  Milestone:
    Component:  metrics-lib  |    Version:
   Resolution:               |   Keywords:
Actual Points:               |  Parent ID:
       Points:               |
-----------------------------+---------------------

Comment (by leeroy):

 I have completed testing of the max descriptor queue size !`feature`
 mentioned in relation to this problem. The size of the queue has no impact
 on onionoo/metrics-lib performance because, in fact, onionoo is limited by
 the rate that parsed descriptors can be produced, which is dominated by
 the blocking call to the parse function. Before providing a fix I need to
 identify what exactly this !`feature` is intended to provide.

 What is the intended use of the setting the max. descriptor queue size?
 Please be specific. Saying it's a feature to limit this size to the rate
 that onionoo is expected to pull the data out makes little sense. Why
 wouldn't you want the pipe as wide as possible on a server?

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/16614#comment:2>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
_______________________________________________
tor-bugs mailing list
tor-bugs@xxxxxxxxxxxxxxxxxxxx
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs