Jump to content

Testing Functionality


Koby

Recommended Posts

It's just we've used 62% of the servers monthly allotted resources in a matter of 13-15 (some sites weren't up at the same time) full days. So I'm going to have to definitely upgrade further next month to improve it. However looking at the logs and such it's likely I can help minimize resource hogs if I can pin-point anything using more than needed.

Link to comment
Share on other sites

It's just we've used 62% of the servers monthly allotted resources in a matter of 13-15 (some sites weren't up at the same time) full days. So I'm going to have to definitely upgrade further next month to improve it. However looking at the logs and such it's likely I can help minimize resource hogs if I can pin-point anything using more than needed.

Wow thanks for all the hard work

Link to comment
Share on other sites

No matter what settings you tweak, it's still likely that, there are just too many people coming and going for your current server plan to handle. This is a big site, as I'm sure the others are too. I'd have to know what things use before I could recommend things to remove. You might be able to just change a few things to make certain features more efficient. Banner changes twice a day instead of every couple of seconds? CB refresh time of 8-10 seconds instead of 5? I don't know all of the features, or what it is about them that eats the bandwidth, so there isn't much else I can say.

Link to comment
Share on other sites

It's just we've used 62% of the servers monthly allotted resources in a matter of 13-15 (some sites weren't up at the same time) full days. So I'm going to have to definitely upgrade further next month to improve it. However looking at the logs and such it's likely I can help minimize resource hogs if I can pin-point anything using more than needed.

i'm suddenly very glad we host our anime on external sources imagine the bandwidth we'd go through if it were all hosted here!

Link to comment
Share on other sites

Thanks for being an awesome dude Koby, the lag was terrible for me personally. It took 2 minutes to load a thread, I didn't have that problem until I could access the chatbox =/. I will be donating more in January in hopes to help with your pain and suffering. Cause I am just plain hooked to rping and drawing on here. heh =)

Hate to throw in this suggestion (cause they are perty to look at) but maybe disabling siggys temporarily will ease some of the loading time?

I wouldn't ever disable them by default. However users are able to disable signatures themselves via 'Edit Options' page under the Thread Display Options category of their User CP. All removing them would do is help load time for those with slower connections.

Ah, I was not aware of this thank you. Has always been a problem for me even though I am running DSL. Maybe its just my inferior connection. Lolz

Link to comment
Share on other sites

Thanks for being an awesome dude Koby, the lag was terrible for me personally. It took 2 minutes to load a thread, I didn't have that problem until I could access the chatbox =/. I will be donating more in January in hopes to help with your pain and suffering. Cause I am just plain hooked to rping and drawing on here. heh =)

Hate to throw in this suggestion (cause they are perty to look at) but maybe disabling siggys temporarily will ease some of the loading time?

Removing the sigs won't do anything, most sigs are hosted on image hosting sites like photobucket. Maybe its just your internet if the sigs load up slow.

Link to comment
Share on other sites

Hate to throw in this suggestion (cause they are perty to look at) but maybe disabling siggys temporarily will ease some of the loading time?

I wouldn't ever disable them by default. However users are able to disable signatures themselves via 'Edit Options' page under the Thread Display Options category of their User CP. All removing them would do is help load time for those with slower connections.

Link to comment
Share on other sites

It'll be the 24 hour mark in half an hour lols! I'm curious to see what the statistics will be.

The load was cut by around 1/3. It went from 14.95 GB (the day before) to 9.94 GB (the day they were disabled) with only about 60 less visitors and 20' date='000 file hits less (which isn't much when you're talking a quarter million).

On EDK I disabled only the rotating banner and it dropped there from 9.51 GB to 6.42 GB with 100 less visitors and 3,000 less file hits (109,521 file hits the day before).

So yes, the difference is pretty big, but yes both sites did get fewer visitors and file hits during the test. However it wasn't a big enough difference to reflect that much of a difference in the results. Sure possibly a 5% difference or so for the forums at most, but even if you add that.. there would have still been nearly a 30% drop in resources used for the day.

I think a lot of it could be the rotating banners. Reason? The way it's done, it's randomly selected and thus they are unable to be cached and therefore every single page load calls for it to freshly load the banner again and sometimes the banners are saved at high quality png of upwards of half a mb in size. Sure that seems like nothing at all, and yeah it really is a small part, but you then have to multiple that by about 80,000 which adds up compared to a cached image that only has to be loaded once and then saved so it doesn't have to be called up until the cache is cleared or the image changed. Sure the numbers are still extremely little for that, but add it up on a daily basis and it becomes a decent chunk of the monthly quota.

[b']*EDIT*

Going to enable only the ChatBox and see how much of a difference it causes now.

Link to comment
Share on other sites

The load was cut by around 1/3. It went from 14.95 GB (the day before) to 9.94 GB (the day they were disabled) with only about 60 less visitors and 20,000 file hits less (which isn't much when you're talking a quarter million).

On EDK I disabled only the rotating banner and it dropped there from 9.51 GB to 6.42 GB with 100 less visitors and 3,000 less file hits (109,521 file hits the day before).

So yes, the difference is pretty big, but yes both sites did get fewer visitors and file hits during the test. However it wasn't a big enough difference to reflect that much of a difference in the results. Sure possibly a 5% difference or so for the forums at most, but even if you add that.. there would have still been nearly a 30% drop in resources used for the day.

I think a lot of it could be the rotating banners. Reason? The way it's done, it's randomly selected and thus they are unable to be cached and therefore every single page load calls for it to freshly load the banner again and sometimes the banners are saved at high quality png of upwards of half a mb in size. Sure that seems like nothing at all, and yeah it really is a small part, but you then have to multiple that by about 80,000 which adds up compared to a cached image that only has to be loaded once and then saved so it doesn't have to be called up until the cache is cleared or the image changed. Sure the numbers are still extremely little for that, but add it up on a daily basis and it becomes a decent chunk of the monthly quota.

is there a way to have say 4 banners in rotation for a quarter of the day each? or does that have to be done manually? otherwise we could change botm to banner of the fortnight and keep the winner up for the two weeks?

also, what about the different themes? i miss my dark skin. all around rpg hurts my eyes

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...
Please Sign In or Sign Up