No, we're not on the new webserver yet, and if it would help, I'd expect its help to be marginable, at best, except that I could safely put more of the load on it and less on the db box. Though the site's performing extremely well for me, and running traces isn't showing anything being misbehaved, it's obvious from the db CPU histogram that the machine is really getting a workout.
I'm waiting for my favorite consultant to take care of setting up the new box, as its problems are going over my head. I'm just a programmer. He's a programmer and a major Windoze hardware/OS guy.
No, Midas, not interested in your new machine unless, perchance, it's got a pair of 3.6Ghz processors, faster and bigger hard drives than we're using now, and at least 4 gig of memory. <g>
The current db is using a pair of 2.8's, so just stepping up .8Ghz on each processor (1.6Ghz total, roughly) would help quite a bit, as would switching to the fastest hard drives money can buy (they're nearly but not quite there) and a RAID configuration that sacrifices some security for the sake of speed.
What I've done so far, and looks to have made an impact, but not good enough for me yet:
1. Disabled auto-refresh of Favorites and Board.
2. Disabled "realtime" (yeah, right -- it was lagging by nearly an hour anyway) full-text indexing and made it an hourly job.
I don't know what I'm going to do next yet. The machine is definitely breathing easier, but it's still not what I'd remotely call "good enough". Was running at about 50% utilization with frequent spikes ot 80%+. Now running from 20-30% with occasional 50%+ spikes. If I can get the spikes below 50%, I'll consider it good enough for now.
I'm hoping the last few percent can be gained my improving the efficiency of current procedures rather than disabling any others.
News 
Market Data 
Discover 
AI
