It's always TCP_NODELAY
brooker.co.za
It's always TCP_NODELAY
1–10 of 276 posts
Re: It's always TCP_NODELAY
#2I don't by the reasoning for never needing Nagle anymore. Sure, telnet isn't a thing today, but I bet there are still plenty of apps which do equivalent of:
write(fd, "Host: ")
write(fd, hostname)
write(fd, "\r\n")
write(fd, "Content-type: ")
etc...
Re: It's always TCP_NODELAY
#3We used to call them "packlets."
His "tinygrams" is pretty good too, but that sort of implies UDP (D -> datagrams)
Re: It's always TCP_NODELAY
#4I don't by the reasoning for never needing Nagle anymore. Sure, telnet isn't a thing today, but I bet there are still plenty of apps which do equivalent of: write(fd, "Host: ") write(fd, hostname) write(fd, "\r\n") write(fd, "Content-type: ") etc... this may not be 40x overhead, but it'd still 5x or so.
Marc addresses that: “That’s going to make some “write every byte” code slower than it would otherwise be, but those applications should be fixed anyway if we care about efficiency.”
Re: It's always TCP_NODELAY
#5I don't by the reasoning for never needing Nagle anymore. Sure, telnet isn't a thing today, but I bet there are still plenty of apps which do equivalent of: write(fd, "Host: ") write(fd, hostname) write(fd, "\r\n") write(fd, "Content-type: ") etc... this may not be 40x overhead, but it'd still 5x or so.
Those aren't the ones you debug, so they won't be seen by OP. Those are the ones you don't need to debug because Nagle saves you.
Re: It's always TCP_NODELAY
#6I don't by the reasoning for never needing Nagle anymore. Sure, telnet isn't a thing today, but I bet there are still plenty of apps which do equivalent of: write(fd, "Host: ") write(fd, hostname) write(fd, "\r\n") write(fd, "Content-type: ") etc... this may not be 40x overhead, but it'd still 5x or so.
The comment about telnet had me wondering what openssh does, and it sets TCP_NODELAY on every connection, even for interactive sessions. (Confirmed by both reading the code and observing behaviour in 'strace').
Re: It's always TCP_NODELAY
#7Not every time. Sometimes it's DNS.
Re: It's always TCP_NODELAY
#8I don't by the reasoning for never needing Nagle anymore. Sure, telnet isn't a thing today, but I bet there are still plenty of apps which do equivalent of: write(fd, "Host: ") write(fd, hostname) write(fd, "\r\n") write(fd, "Content-type: ") etc... this may not be 40x overhead, but it'd still 5x or so.
Fix the apps. Nobody expect magical perf if you do that when writing to files, even though the OS also has its own buffers. There is no reason to expect otherwise when writing to a socket and actually nagle already doesn't save you from syscall overhead.
Re: It's always TCP_NODELAY
#9I don't by the reasoning for never needing Nagle anymore. Sure, telnet isn't a thing today, but I bet there are still plenty of apps which do equivalent of: write(fd, "Host: ") write(fd, hostname) write(fd, "\r\n") write(fd, "Content-type: ") etc... this may not be 40x overhead, but it'd still 5x or so.
And they really shouldn't do this. Even disregarding the network aspect of it, this is still bad for performance because syscalls are kinda expensive.
Re: It's always TCP_NODELAY
#10I don't by the reasoning for never needing Nagle anymore. Sure, telnet isn't a thing today, but I bet there are still plenty of apps which do equivalent of: write(fd, "Host: ") write(fd, hostname) write(fd, "\r\n") write(fd, "Content-type: ") etc... this may not be 40x overhead, but it'd still 5x or so.
The comment about telnet had me wondering what openssh does, and it sets TCP_NODELAY on every connection, even for interactive sessions. (Confirmed by both reading the code and observing behaviour in 'strace').
Especially for interactive sessions, it absolutely should! :)