Re: [PATCH v2] fetch-pack: grow stateless RPC windows exponentially

From: Junio C Hamano <gitster@pobox.com>
Date: Jul 19, 2016, 19:23 UTC
Message-ID: <xmqq8twxfn4j.fsf@gitster.mtv.corp.google.com>
In-Reply-To: <CAGf8dgL3t7uX7yAux0xc2QMJJdmnM0262Quj4o6gDehwA+4JqQ@mail.gmail.com>

Jonathan Tan <jonathantanmy@google.com> writes:

Show 7 quoted lines

>> So I guess what I do not understand is why we need to slow down the
>> exponential growth at all?
>
> The current code has an exponential (a' = a * 2) then a linear (a' = a
> + 1024) growth. I'm not slowing down the exponential growth - that
> part is retained. I'm replacing the linear growth with another
> conservative exponential growth (a' = a * 11 / 10).

As stateless-rpc mode is to drive a half-duplex channel, the function essentially determines how many messages to buffer before passing the control to the other side. The increment between number the function is called with and the function returns is how much the other side is made to wait, i.e. how long the ping-pong latency is.

Even if it is conservative, I wonder if it is truly a good idea to make it exponentially grow forever from that point of view. Would it give essentially the same result to you if we discard the patch in question and just raise LARGE_FLUSH to 10k instead?