"And the sql query won't be fast if there are are too many records."
I don't think this is true. The datastore speed is designed to vary with
the size of the result set, not the size of the data set.
On 23 February 2012 00:58, Kesava Neeli wrote:
> our user account has many fields like userName
you could create a simple list of strings like "googleId:2182031",
"facebookId:2913812" and query it on a single field depending on your needs
On Wed, Feb 22, 2012 at 10:58 PM, Kesava Neeli wrote:
> our user account has many fields like userName, email, number, alternate
> email list, ids from
our user account has many fields like userName, email, number, alternate
email list, ids from social networks like facebook, google, twitter,
linkedin etc. The primary key is one of those fields. In "realtime", we
should be able to find if a user exists with any the fields and not just
email. O
Your requirement appears to be "able to look up user by email". The
datastore does exactly this for you. The idea of loading your entire
datastore into memcache just so you can do lookups is totally absurd.
Jeff
On Wed, Feb 22, 2012 at 6:05 PM, Kesava Neeli wrote:
> Jeff,
> If you understood
Jeff,
If you understood my requirement, I would love to hear any suggestions you
have if you think MemCache is not a solution.
--
You received this message because you are subscribed to the Google Groups
"Google App Engine for Java" group.
To view this discussion on the web visit
https://gro
You are overthinking this. To the point of lunacy.
Jeff
On Wed, Feb 22, 2012 at 5:07 PM, Kesava Neeli wrote:
> We would like to have a map of all users in datastore in mem cache so that
> we can look up for any username/email/number to verify if he is already an
> user of our app. Ex: A user w
We would like to have a map of all users in datastore in mem cache so that
we can look up for any username/email/number to verify if he is already an
user of our app. Ex: A user wants to send some message to an email address
from our mobile app. We want to find if there is any user in our databa
Are all of these 1-2 million users going to be active at the same time? If
not, why do you care if it all fits in memcache?
On the other hand, if they *are* all going to be active at once, I would be
very curious to know what happens...
Jeff
On Tue, Feb 21, 2012 at 2:07 PM, Kesava Neeli wrote:
Hi IKai,
Could you point me to the docs about the max memory for the app for
MemoryCache? We have a mobile app with many user accounts and each account
has a primary key. We expect to have 1-2million accounts in the datastore.
We need to store many entries in mem cache to find if account exist
There isn't a limit per se (that I know of), but our implementation of
Memcache has two LRU (least recently used) queues that are used to expire
content based on recency of access: a global LRU for that particular
Memcache instance, and an LRU for your app. The amount of memory used is
subject to c
Is there any limit of maximum keys i can store in memcache in any
particular time. Can store billions of keys as long as each key not
exist hardlimit 1 mb? Is there any exacts number?
--
You received this message because you are subscribed to the Google Groups
"Google App Engine for Java" group.
11 matches
Mail list logo
