These aren't the kind of quantitative I meant. ;) 10 times nothing is still nothing. What kind of overhead, in terms of time and CPU usage in a typical request in your application? It's just a bunch of string manipulation, it should be really cheap compared to even a single network/memcache/datastore access (seeing as CPU quota seems to be based mainly on request duration).
2009/3/24 Andy Freeman <[email protected]>: > > The referenced post has examples of simplejson taking 10x more time > than cjson. 10x difference in a component is noticeable for some > applications. > > It also shows that pickle is much slower and has 2x bigger output than > json for "objects". (For strings, they all should be roughly the > same.) > > I mention pickle because it's used by memcache. Since both pickle and > json should be very efficient for strings, it may make sense to > memcache json output instead of objects. > > In addition, some people are using pickle to create datastore blobs. > Perhaps json is a better choice. > > On Mar 24, 1:39 pm, David Wilson <[email protected]> wrote: >> Just idly wondering, >> >> Have you done any quantitative measurement of how "bad" simplejson is? >> I can't imagine it being a large overhead, unless your application is >> seriously optimized to extremes already. >> >> David >> >> 2009/3/24 Andy Freeman <[email protected]>: >> >> >> >> >> >> >> >> > That's why issue 1174 requests the fastest possible json, not cjson. >> >> > 1174 actually requests multiple jsons, so folks can use what's best >> > for their application. >> >> > On Mar 23, 10:32 pm, Joseph Turian <[email protected]> wrote: >> >> I am the author of the blog post. >> >> >> <b>Update (20090324):</b> According to <a href="http:// >> >> kbyanc.blogspot.com/2007/07/python-serializer-benchmarks.html" >> >> rel="nofollow">Extra Cheese</a>, cjson 1.0.5 has an incompatibility >> >> with simplejson in processing slashes. A fix is available from <a >> >> href="http://www.vazor.com/cjson.html"; rel="nofollow">Matt >> >> Billenstein</a>. However, Dan Pascu, the author of cjson, deprecates >> >> Matt Billenstein's cjson 1.0.6 because Matt's patch parses the JSON >> >> twice, which makes it twice as slow. This will still be faster than >> >> all alternatives in certain circumstances. You will not find Matt's >> >> cjson on the cheeseshop, only on Matt's site. >> >> >> On Mar 23, 10:55 am, Andy Freeman <[email protected]> wrote: >> >> >> > According >> >> > tohttp://blog.metaoptimize.com/2009/03/22/fast-deserialization-in-python/ >> >> > , simplejson is significantly slower than cjson. >> >> >> > I've created an issue requesting the fastest possible json >> >> > athttp://code.google.com/p/googleappengine/issues/detail?id=1174.-Hide >> >> > quoted text - >> >> >> - Show quoted text - >> >> -- >> It is better to be wrong than to be vague. >> — Freeman Dyson- Hide quoted text - >> >> - Show quoted text - > > > -- It is better to be wrong than to be vague. — Freeman Dyson --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Google App Engine" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/google-appengine?hl=en -~----------~----~----~----~------~----~------~--~---
