I've been investigating the appengine to see if I can use it for a project and while trying to choose between Python and Java, I ran into a surpris开发者_运维知识库ing difference in datastore query performance: medium to large datastore queries are more than 3 times slower in Python than in Java.

My question is: is this performance difference for datastore queries (Python 3x slower than Java) normal, or am I doing something wrong in my Python code that's messing with the numbers?

My entity looks like this:

Person

firstname (length 8) lastname (length 8) address (20) city (10) state (2) zip (5)

I populate the datastore with 2000 Person records, with each field exactly the length noted here, all filled with random data and with no fields indexed (just so the inserts go faster).

I then query 1k Person records from Python (no filters, no ordering):

q = datastore.Query("Person")
objects = list(q.Get(1000))

And 1k Person records from Java (likewise no filters, no ordering):

DatastoreService ds = DatastoreServiceFactory.getDatastoreService();
Query q = new Query("Person");
PreparedQuery pq = ds.prepare(q);
// Force the query to run and return objects so we can be sure
// we've timed a full query.
List<Entity> entityList = new ArrayList<Entity>(pq.asList(withLimit(1000)));

With this code, the Java code returns results in ~200ms; the Python code takes much longer, averaging >700ms. Both apps are on the same app id (with different versions), so they use the same datastore and should be on a level playing field.

All my code is available here, in case I've missed any details:

http://github.com/greensnark/appenginedatastoretest

This would be an expected difference between Python and Java. Most likely you aren't seeing differences in the amount of time to make the query, but the amount of time it takes to parse the result and fill the receiving data structure.

You can test this by comparing the time it takes to query a single record. Remember that you'll need to test several times and average the total to get a true benchmark to account for possible fluctuations in latency on the backend.

In general, you can expect a compiled statically typed language like Java or Scala to always be faster than an interpreted language dynamically typed language like Ruby or Python.