Hacker News Reader

Comment by technofiend | Hacker News Reader

Oh, God. Are you serious? Listen try sitting in Texas some time and opening a X application even over a compressed SSH link from Delaware. It literally takes minutes for a splash screen to paint. And using applications are nearly impossible due to interaction lag. You're better off using a VMware console, RDP heck even Citrix which at one time was itself supposedly X based.

X is simply awful over the WAN, trust me. Please note: I'm not talking about an essentially text-only experience like xterm, which works fine. I mean actually using X as a graphical client.

View on HN

The problem there is not so much X, as software devs not using X but instead just painting everything into "image" that is then dumped into X. This means you are effectively looking at what amounts of a VNC session, without the benefit of compression.

There are at least one X extension out there that offer compression for long distance connections. But for some reason they never really took off.

I did mention compression of the X session itself by way of compressing the SSH tunnel. The problems run deeper than "software devs just painting everthing into image". X itself has a mix of synchronous and asynchronous calls that are very sensitive to latency. XCB instead Xlib and use of LBX can help but do not eliminate the performance issues with X Window.

See http://keithp.com/~keithp/talks/usenix2003/html/net.html and http://vis.lbl.gov/Events/SC08/RemoteX/index.html for more details.

As it relates to the original topic I'd be interested to see real-world performance comparisons between Plan 9 and X because as the papers state X stinks on ice over the WAN.

Does plan9 works well over a WAN? (Genuine question)

Yes. For remote control, drawterm (a client to "cpu" into a plan9 box from a non-plan9 box) can even make the GUI feel local over poor connections (except if you try to play doom or something).

That's over a LAN. But what about over the distances earlier in the thread - the earlier example was a line across a significant bit of continental US. [More interesting still would be Sydney to London.]

No, I was talking about WAN, not LAN. (Note that 9front's fork of drawterm is a lot faster than the normal version - a regular drawterm can be a bit slow)

I am not located in the US, but hopping borders in europe does not seem to be a problem. It's less of a problem than with X-forwarding or VNC.

As for regular fs mounts (note that a remote control session on plan9 is actually also just a regular fs mount), I mount things from the US every once in a while, and I am located in Sweden. It's obviously not fast due to very high latency and limited bandwidth, but it works just fine within those constraints.

But of course, if the RTT >1 second, then it will take >1 second for a keystroke to show up on screen in a remote control session. You cannot be resistant to this (mosh tries to guess how things would look if the keystrokes had gone through, which is a major hack, and not really an issue).

Nomachine worked reasonably well for me cities apart. I believe it was adding some compression to X at the time; circa 2011.

Reply on news.ycombinator.com