DataAccessPerformance
DataAccessPerformance copied to clipboard
Benchmarks, prototypes and discussions for database access performance investigation
Attendees: @ajcvickers @anpete @halter73, @roji @sebastienros @divega Today @halter73 gave us a great introduction on the history and future of optimizing Kestrel and how some of the same ideas could...
Attendees: @roji (he joined us in person today!) @anpete, @ajcvickers, @sebastienros, @mikeharder, @divega) Ideas of things that we can investigate in the short term 1. Profiling Npgsql and Peregrine and...
_Attendees: @ajcvickers @anpete @divega @geleems @sebastienros_ @sebastienros gave us an update on the progress on Npgsql at https://github.com/aspnet/DataAccessPerformance/issues/5#issuecomment-349080515. The latest experiment shows very promising results. @sebastienros also gave us on...
_Attendees: @ajcvickers, @DamianEdwards, @geleems, @divega, @sebastienros_ - We have a repo! @sebastienros is going to initialize it with the console benchmark code an .gitignore file (we may want to take...
Currently we have a limited set of repeatable benchmarks and we are using assumptions to extrapolate how our results compare with other results obtained from different environments. The more we...
We have some ideas that we want to try, which involve giving low-level networking for ADO.NET providers a similar makeover treatment to what did with Kestrel some time ago. @davidfowl...
@damageboy, @roji and @anpete have been discussing and looking possible changes to Npgsql as well as modifying the benchmark code to try to get better performance(see https://github.com/npgsql/npgsql/commit/c153d2772f9392851922e87b3f989031a152410b, [insert a reference...
Some initial thoughts; not sure if I should log these individually or whether this is fine, but: - not sure how well it works for non-query and multi-grid query scenarios...
This is something that came up in brainstorming and is related to https://github.com/aspnet/DataAccessPerformance/issues/4: The assumption is that if it is true that ADO.NET has fundamental design issues then it should...
We have had some discussions about ADO.NET APIs possibly containing fundamental design issues that would prevent ever achieving the best performance. For now just creating a placeholder issue here: we...