How should I test 'your approach'? Would you share a patch? Please, compare the two tests with the same number of iterations.
Paolo On Wednesday, November 20, 2013 12:04:35 PM UTC+1, Arnon Marcus wrote: > > I get these results: > > C:\Python26\python.exe F:/Experimental/web2py_272/web2py/test_executesql.py >> Times are for 100 iterations >> Preparing.... >> started bench >> speed_executesql() total=1.132703 avg=0.011327 >> speed_normal_rows() total=5.320367 avg=0.053204 >> speed_rows_with_cacheable() total=5.349665 avg=0.053497 >> speed_executesql_with_colnames() total=1.083485 avg=0.010835 >> speed_rows_with_simple_processor() total=1.202628 avg=0.012026 >> Process finished with exit code 0 > > > Pretty similar results... > Although a little slower overall, same gaps appear. > > > Then I tested it with timeit using your approach: > > Times are for 10 iterations >> Preparing.... >> started bench >> speed_normal_rows() total=47.711369 avg=4.771137 >> speed_executesql_with_colnames() total=0.513249 avg=0.051325 > > > That is a *x92.959318 times* (!) faster > Almost full 2 orders of magnitude. > > > On Tuesday, November 19, 2013 10:45:52 PM UTC+2, Niphlod wrote: >> >> if only those announcements came with actual reproducible tests..... life >> would be easier for us :-P >> I'd guess your usecase is either 1k rows with 50 columns or a really slow >> python server, because I can't reproduce those kind of "advertised >> magnitude gains". Just a 4x that is quite expected since Rows carries >> around lots of helpers. >> >> CPython 2.7.3 >> speed_executesql() total=0.846748 avg=0.008467 >> speed_normal_rows() total=4.214647 avg=0.042146 >> speed_rows_with_cacheable() total=3.705196 avg=0.037052 >> speed_executesql_with_colnames() total=0.844287 avg=0.008443 >> speed_rows_with_simple_processor() total=0.891577 avg=0.008916 >> >> pypy 1.9.0 >> speed_executesql() total=3.344496 avg=0.033445 >> speed_normal_rows() total=3.832903 avg=0.038329 >> speed_rows_with_cacheable() total=3.472066 avg=0.034721 >> speed_executesql_with_colnames() total=2.808536 avg=0.028085 >> speed_rows_with_simple_processor() total=3.268108 avg=0.032681 >> >> Attached the test code. >> >> -- Resources: - http://web2py.com - http://web2py.com/book (Documentation) - http://github.com/web2py/web2py (Source code) - https://code.google.com/p/web2py/issues/list (Report Issues) --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
