QuickBase Performance
I wasn’t sure if the difference was real, or just my weekend brain running slow, so we put together a benchmark script to see how QuickBase performs hour-to-hour and day-to-day. A QuickBase app runs the script twice per hour, logging the date, time, and duration when the script ran. An increase in script execution time corresponds with a decrease in QuickBase’s responsiveness.Here’s what we found.
The weekday performance graphs are surprisingly symmetrical. Every morning, from about 6 to 9 (all times are Eastern), the system moves very quickly - our benchmark took anywhere from 8 to 11 minutes to run. This is followed by a sharp rise in response time, to anywhere from 10 to 18 minutes, which lasts until about 6 PM. The numbers fall again from 6 to 9 PM. In the middle of the night, there is another spike, most likely as scheduled batch jobs run.
QuickBase performace on Weekdays
 |
So, for maximum weekday performance, it seems that you’d be best off fitting your work in when you’d ordinarily be eating breakfast or dinner.
Weekends are a completely different (and rather more baffling) story. Saturday and Sunday both show the worst performance times between the hours of 9 PM and 3 AM. The rest of the day is relatively constant at a substantially lower time.
QuickBase performance, Weekdays vs Weekends
 |
Bottom Line, I was pleased at how consistent QuickBase’s performance is. There will always be some slowdown during peak work times, but performance never is worse that 50% of off-peak, which I think reflects a good job of load balancing on QuickBase’s end.
This entry was posted
on Thursday, June 10th, 2010 at 12:13 pm and is filed under Articles.
You can follow any responses to this entry through the RSS 2.0 feed.
You can leave a response, or trackback from your own site.