|
|
Synopsis
A major gaming company used eValid functional test monitoring services
to analyze the time customers need to download their medium-sized
(10 MByte) application.
After several months they were able to make server adjustments that
decreased total download times and minimized the variance in performance
their users has been noting.
Background
The on-line gaming company relies upon download of a special-purpose
game module, which is downloaded to the user's local machine and
which, in operation, interacts with the company website as a user "plays".
Security is a major issue since real money is involved, and to support
enhanced security the download of the client is repeated fairly often.
Also, in some cases different games have different clients.
The company began to realize that, as the complexity of the game client grew, so did its size and so did the number of users who declined to wait for the download to complete. Not every download server behaved the same, and the problem was to find out which download server did the best job and to balance the requested downloads between all available servers for the best e-commerce effect.
eValid Application Description
To complete the measurement eValid engineers developed a simple
script that times the FTP download using internal image-based
synchronization.
The script would download the target "game client", time how long it
took to say "Download Complete", and report the results to the timing logs.
This script (and several variations of it, one for each separate game being timed) was put in 15-minute monitoring intervals operation and data was collected over a multiple-month period. This data was used to balance the servers to better match the demand.
Customer Comments
"We're out of the woods on this one, and we owe it all to
eValid. Without the timing data we would never have found the
right balance among all of our servers." --Mark K.
