Yep the blobstore approach is a better solution IMHO but costs will
still be prohibitive. Im afraid
nothong much you can do about that. Initially i thought that we could
get an export of the datastore, download it as a binary file and
process it locally (like the local .bin used for devs) but i'm
dreaming out loud :-)

On Dec 30, 4:55 am, Jeff Schnitzer <[email protected]> wrote:
> On Thu, Dec 29, 2011 at 2:58 AM, Brandon Wirtz <[email protected]> wrote:
>
> > I would say GAE handles big data really well. But you have to do testing to
> > make sure your structure is correct, and that your indexes are well thought
> > out.
>
> I think we are talking about two different things.  I'm thinking of
> Big Data like this:
>
> http://en.wikipedia.org/wiki/Big_data
>
> Typically characterized by:
>
>  * Large data volumes
>  * Batch updates
>  * Frequent need to analyze/sift through large quantities of data
>
> The GAE datastore performs poorly in this regard.  Map/reduce support
> is anemic at best.  Per-gigabyte storage is expensive.  Raw I/O
> performance is *dreadful*.  Indexes consume excessive amounts of
> space.
>
> I love the GAE datastore, I think it's hands-down the Best Storage
> Around for web applications that need scalability and availability.
> But there's no way in hell I would use it to store a large-scale OLAP
> system or any other kind of serious analytics product.  You don't want
> EC2 either.  You need something like Hadoop on bare metal hardware
> with really fat I/O pipes.  It will cost you a tiny fraction of what
> you'll spend at Google will cost and perform 10X better.
>
> Jeff

-- 
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