Read can be cached in memory, speed is about 4GB/s. If your entity is
about 100kb, you can do 40,000 get() in a second.

A put() needs about 3 writes and 1 read, each writes needs at least
10ms seek time. Considering the network latency and disk write time,
most entities cannot be written more than 5 times in a second.

2010/2/3 marcdmarc <[email protected]>:
> Hello,
>
> I am trying to get clarification on a statistic given by Brett
> Slatkin's google i/o talk on building scalable apps on app engine.
> This is the talk I am referencing.
> http://sites.google.com/site/io/building-scalable-web-applications-with-google-app-engine
>
> In the discussion, Brett mentions writes have a rate of 100 seeks per
> second where the size and shape of your data actually determining the
> write throughput of an entity.  He then goes on to say that "You can
> think of this as being the maximum write throughput for a single
> entity on disk."
>
> To me, this statement means that a single entity can be written up to
> a maximum of 100 times per second.  I would like to clarify that my
> interpretation of Brett's statement is correct.
>
> If anyone could clarify my understanding to be correct, I would
> appreciate any input that people have.
>
> Best,
>
> Marc
>
> --
> 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