Dr. Hipp, > Ulrik Petersen wrote: >> >> has anyone experienced less performance with SQLite 3.0.7 over 2.8.13 on >> the same data? That is what I am experiencing. I'd appreciate help in >> figuring out why and perhaps what I can do about it. >> > > SQLite 3.0 requires less disk I/O at the expense of using more CPU cycles. > So if you have a fast disk and a slow CPU, SQLite 3.0 might well be > slower. On the other hand, a slow disk connected to a fast CPU will > make SQLite 3.0 faster. On my 3-year-old Athlon with an 7200RPM > IDE disk, SQLite 2.8 and 3.0 are about the same speed. But I figured > that CPUs tend to increase in speed more rapidly that disk drives, so > it was best to optimize for a faster CPU. > > Might this explain the result you are seeing? Do you (perhaps) have an > older CPU and/or an exceptionally fast disk drive?
Thanks for the information. I have an AMD Athlon 3000+ with a 7200RPM IDE disk running at ATA100, so I am not sure whether that explains it. I may try it on one of my other computers and see what benchmark results I can get. Does SQLite 3 take longer to parse the schema at startup? The benchmark queries I run are run in strict, non-overlapping sequence, with a full sqlite3_open and sqlite3_close in separate processes. Yet even when I concatenate the queries and run them all in one go, SQLite 2.8 is faster than SQLite 3. Thanks in advance. Ulrik
