True, but there is also CPU overhead for the two rpcs.
On Jan 5, 2012 6:02 PM, "Jason Collins" <[email protected]> wrote:

> There is still $$ difference. Memcache is (or appears to be) free,
> datastore reads are not.
> j
>
> On Jan 5, 8:01 am, Andrin von Rechenberg <[email protected]> wrote:
> > Hey there
> >
> > For latency reasons we cache frequently fetched datastore entities:
> http://code.google.com/appengine/articles/scaling/memcache.html#entit...
> >
> > *def getProfile(id):*
> > *  profile = memcache.get(id)*
> > *  if not profile:*
> > *    profile = datastore.get(id)*
> > *    memcache.put(profile)*
> > *  return profile*
> > *
> > *
> > So we first check if we can get it from memcache,
> > if not we get it from datastore and put it into memcache.
> > Simple. In many cases the a profile is not in memcache,
> > because we have a lot of profiles. So we do 3 RPCs.
> >
> > Looking a bit at appstats it seems that these days a
> > datastore.get_by_key_name is as fast as a memcache.get
> > in many cases, so I'm wondering if google has put a
> > memcache infront of datastore and the pattern I'm doing
> > above is obsolete and I could save the extra RPCs?
> >
> > Or maybe datastore is just having a good day?
> >
> > Cheers,
> > -Andrin
>
> --
> 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.
>
>

-- 
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.

Reply via email to