Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Mac Kerr]

Ernie, i don't know how much you follow all the forums, but there is an ongoing thread in the Classic LAB about the recent slowdown in system response. Thanks

Mac

[Dave Nolan]

Link to message - Just in case you can't be bothered finding it and want to read it...

Dave M.

[Mike MacWillie]

Seems to be running fine today.

[Ernie Black]

Sorry guys, I had the nerve to take a brief vacation this past week--and have been paying for it ever since! Anyway, here's my response to the orginal thread:

kinnaman wrote on Thu, 19 August 2004 15:36

First, props to the gang that keeps this site running!

Thanks, it's always nice to get our props (even if we know chops are to follow   )

Quote:

Consider that every post gets fed into a keyword-searchable database (which is processing-intensive) and that a staggering number of people are trying to simultaneouly access the same information, and I would have to say that this incarnation of the LAB, though not perfect as fast as we would like it to be, is an incredible step forward.

Very good point on the database processing. However, there are any number of factors which can and at times do impact the delivered speed of the forums. At the bottom of each forum page you should see a readout of how long it took the system to generate the page you're currently viewing (e.g., Total time taken to generate the page: 0.17086 seconds). Other factors involve not only the download speed at your end (dial-up modem, cable, dsl, etc.) but also the hops between you and us (the "tight string between the tin cans"--nicely put) which you can always check by running a traceroute command to the server (e.g., in a Windows command prompt, type tracert and the domain you're trying to reach, such as srforums.prosoundweb.com, and you'll get a step-by-step display of the hops and the time spent for each hop between here and there. Since we can't afford multiple mirrored servers scattered across the net a la Akamai, the route betwee you and us can make a huge difference.

Other factors can include the number of users and the type of activity (lots of searches at once, for instance), thus the level of server activity at any given time. Today things appear, to me at least, to be flying rapidly (I'm on a broadband cable modem, 17 hops between me and the server with transient time running between 11 and 35 ms between hops). Of course, other bits (like the Google ads and our own ads) add to the download time as well; I doubt that I need comment on the need for ad revenue to keep the place running.

Cheers!

[Tim McCulloch]

Ernie-

Back in the "old days" I'd have run tracert, but any more....

It *was* slooooooow.  Dial up speeds on my DSL connection, just like the other guys said.... and it was just as slow on my neighbor's cable connection... Hmmm, Cox and SBC both get the same results.  They seem to have very different routing tables IIRC (I did do some traces when I got dsl last year) so unless there was a backbone problem somewhere, I was suspecting a server problem on the PSW end of things.

At any rate, it's cleared up now, and all is good.

BTW, my question about Chuck's 'Driving Looong Lines" article had to do with finding it from the Study Hall links... I still can't find it in *any* of the categories offered on the study hall page.  Yes, I know it's still of the server, and you and another LABster pointed me to it.  So tell me, please, exactly *where* the link lives?  That's why I posted, couldn't find it from the SH pages.... Not on the "Cabling and Interconnect" page, not on "Audio Basics," no link on the "SR Techniques" page... so where the heck is this thing hiding?

Great job, Ernie, thanks for holding things together for us!

Tim Mc

[Ernie Black]

Tim,

As I *think* I stated, among the many variables which could have affected the delivery speed of the site could have been activity on our forums server, a level up from the server at our soho, a level up from that at our colo's backbone to the 'net, or somewhere else between there and there (us and you). I don't know the cause at the time, and it seems speedy enough now, so there's no way of really telling.

As for Chuck's 'Driving Looong Lines' article, the easiest way to find this (or any SH article) would be to use the PSW search: plug in the title or phrase and it should be found. In this instance, no results are returned for driving looong lines because you're missing an /o/ in Looo[o]ng: Chuck used four o's, not three, for emphasis. Or search for Driving Long Lines with only one /o/ and the article comes up (it's physical location is http://www.prosoundweb.com/studyhall/lastudyhall/longline.ht ml).
As for it not being in the one of the SH sections, that's likely true, as it (to my knowledge) was never reformatted to an actual PSW webpage from its original LAB post. It seems that it should be, and it should be linked from "Cabling and Interconnect", so I'll create the appropriate page for it.

Just shows how much we need to maintain the old LAB archives! None of our other archives get anywhere near the traffic that the LAB does.