[google-appengine] Re: Datastore Read and Writes Statstics

2011-11-13 Thread Tom Fishman
Is there an update on this? Have we filed an issue? Our quota depleted so fast when we use map-reduce... On Sep 29, 7:19 pm, djidjadji djidja...@gmail.com wrote: It would be really nice if the WriteOps and ReadOps are logged for each request. Like the loading_request and ms time that is added

[google-appengine] Re: Datastore Read and Writes Statstics

2011-09-22 Thread Vivek Puri
Rishi Thank you for your suggestion. I have been using it. But for a large app, it is a manual way of finding out which table is getting hit most. Also, if tables have few fields with indexed=false, it becomes really hard to figure out the true cost of writes across tables. On Sep 22, 8:57 am,

[google-appengine] Re: Datastore Read and Writes Statstics

2011-09-22 Thread Vivek Puri
As per this doc: http://code.google.com/appengine/kb/postpreviewpricing.html#operations_charged_for There are 3 categories of Datastore operations: Write operations (Entity Put, Entity Delete, Index Write) Each of these operations will cost $0.10 per 100k operations. Read operations (Query,

[google-appengine] Re: Datastore Read and Writes Statstics

2011-09-22 Thread Joops
Just to jump in here, I also agree that we could do with an easy way of identifying where costs are coming from. I like appstats, but it gives you a detail level view for one request. Perhaps information like this might be useful for people Entity Name - Total Writes Total Reads

Re: [google-appengine] Re: Datastore Read and Writes Statstics

2011-09-22 Thread Rishi Arora
I'll take a shot at mapping these. datastore_v3.Put will result in one Entity Put + One Index Writes for each index field that was added (in case of new records), or Two Index Writes for each index field that was modified (an index delete for the old index value and an index write for the new