[Gllug] Excuses for slow server response times

[Simon Perry]

Hi All,

I use a web service API that in it's first stage simply returns results 
from a database. On Xmas eve execution times went from a .5 second 
average to a 6 second average and stayed that way until I finally got 
hold of the company in early Jan. Since then I have been badgering for 
an explanation and assurance that it won't happen again. Now I don't 
know what platforms / databases they run so this will be very general 
conjecture.

I have received a reply today that I'm pretty dubious about. I don't 
want to quote directly but to summarise

Response time increased because less people were using the service over 
that period.

Less searches were stored in memory so results were retrieved from the 
database and were thus slower.

Even if they employ database caching some queries must have been 
duplicated over that time. Yet no call was quicker than 5.5 seconds with 
the slowest being a massive 68 seconds.

What do you think?

Simon


-- 
Gllug mailing list  -  Gllug at gllug.org.uk
http://lists.gllug.org.uk/mailman/listinfo/gllug