From: Alex Young <alex@...>
Date: 2006-08-06T16:48:21+09:00
Subject: Re: [ANN] RFuzz 0.7 -- Win32, Bug Fix, Browser Begins

Alex Young wrote:
> Zed Shaw wrote:
<snip>
>> I'd also be curious for the same test run with the RFuzz call being done
>> inside a Timeout block.  RFuzz doesn't do any of that (bare metal), and
>> I know Timeout blocks fire up an extra thread and eat up a bit more CPU.
>> I'm betting that putting RFuzz inside Timeout makes it about the same
>> speed at net/http.
> 
> Not quite:
<snip>
> But then again, that's not what Net::HTTP's actually doing - it puts a 
> timeout around every single rbuf_fill call, so by default it has to 
> spawn a new thread for every 1K of data it intends to receive.  I'm sure 
> there's a damn good reason for that, but it's a reason I don't think I'd 
> have spotted if I was writing it. 
Ok, having actually stopped and thought about it for a second, I can see 
why they've done that.

> That shouldn't be a problem on this 
> test, though - the file it's downloading is only 715 bytes.  I'll try 
> the test with a bigger file and see what I can see.
> 
alex@pandora:~/Desktop/Noodling/ruby$ ruby rfuzz_test.rb
Rehearsal -------------------------------------------------------
net/http              7.890000   1.490000   9.380000 (  9.411633)
net/http big        1055.150000 119.690000 1174.840000 (1185.696163)
rfuzz                 2.840000   0.990000   3.830000 (  4.065557)
rfuzz+timeout         3.380000   1.190000   4.570000 (  4.569331)
rfuzz big           169.730000  47.670000 217.400000 (218.332456)
rfuzz+timeout big   205.470000  59.000000 264.470000 (271.745557)
------------------------------------------- total: 1674.490000sec

                           user     system      total        real
net/http              7.880000   1.470000   9.350000 (  9.454987)
net/http big        1056.710000 113.930000 1170.640000 (1173.337412)
rfuzz                 2.850000   0.980000   3.830000 (  3.831175)
rfuzz+timeout         3.420000   1.180000   4.600000 (  4.609152)
rfuzz big           175.550000  45.900000 221.450000 (221.584886)
rfuzz+timeout big   205.930000  55.530000 261.460000 (262.673411)

The .*big results are from downloading the results of dd if=/dev/zero 
of=big_file bs=1M count=1...  that is, 1MB of zeros.  Interesting.

-- 
Alex