I'm thinking about developing a RESTful web service that will return lots of data. Exactly what do people think may be the maximum size xml document that needs to be came back in a single get from the web services API? I'd estimate how big the end result set to become 100Mb and time for it could be about 2h.
The length of time is really a reasonable maximum for that server to become calculating an effect set before coming back the document?
The net service would simply be known as a couple of times each day.
What are the special configuration options I ought to enable to make this happen around the Apache/mod_python/django stack.
This is actually a constraint of HTTP and absolutely nothing related to Relaxation. Personally, I regularly download content that's 100s of mb over HTTP and haven't any problems.
In case your web service is made to server large documents, then don't be concerned in regards to a maximum. However, should you serve a mixture of big and small documents, you might like to consider serving some type of summary, together with a URL fully document. Your primary concern ought to be to properly configure the net server additional since serving a lengthy document ties up an association for a longer period period. The particular service process is not so CPU intensive though.
The Relaxation model doesn't impose size restrictions, and many people are utilizing it for everyone up .ISO images of Compact disks and Dvd disks. The DVD images are 2-6 gb.
The utmost dimensions are 50 if you do not set a default. Should you choose, it seems to become a limit of 1000 that you could set.
This appears to become consistent among multiple uses, so you might take that like a convention.
I'd say perform some testing, but when it's only two times daily, go for that maximum and come lower for those who haveOrwish to.