| |
|
Thursday, April 18, 2002
|
|
Here's another article regarding SOAP performance that articulates the long version of my previous post. [gtb]
3:09:52 PM 
|
|
While it may be true that the SOAP protocol is "fat" and wasteful of bandwidth, that is probably not the key problem application developers will encounter with respect to performance. Network latency, tunneling through firewalls, moving from the public Internet through application servers and hosts via a SOAP messaging layer will all take time. If you expect an application to make too many of these calls too often, your application will be slow. If the granularity of these SOAP calls is at the accessor method level, your application will be too slow. If it is executed once an hour to synchnronize a batch of information that is worth the overhead of a call, your SOAP application performance will be just fine, thank you very much. [gtb]
3:08:58 PM
|
|
I was asked about my thoughts on the performance of SOAP-based applications. My immediate reaction was "they'll be pigs". After thinking about it a bit longer, I realized that the correct answer is probably, "depends how you use it". In looking for real world evidence to support either view, I came across this article from this past winter: "You see, there's an evil little secret about Web services that most vendors don't talk about. Web services' protocols are very fat, and that means that Web services interactions over the network will be slow and eat up a large chunk of bandwidth." [gtb]
3:04:38 PM
|
|
An excellent technical brief on SOAP by Clay Shirky. These O'Reilly briefs are outstanding. They really get the root of the key practical technical issues. Just like their books.[gtb]
2:37:56 PM
|
|
Bell Mobility is hosting a Wireless Internet conference here in Toronto next week. A one day affair, the agenda looks pretty solid. [gtb]
2:12:33 PM
|
|
|
© Copyright 2002 Intelliware Development Inc..
Last update: 5/23/2002; 4:23:32 PM.
|
|
|
