ANTS Performance Profiler feature suggestions
Thanks for visiting the feature suggestions page for ANTS Performance Profiler from Red Gate! We’d love to hear your ideas for new or improved features that would make ANTS Performance Profiler more useful to you.
Got a query about current functionality? Please visit the Support Center, or post your question on the forum.
We read and will consider all of your feedback – thank you for sharing your ideas!
-
Log exception message + callstack
Currently, only exception types get logged. It would be very helpful to have the message and callstacks at hand. Clicking the exception details could automatically navigate to the source code position where the exception was thrown.
1 vote -
Add support for profiling Postgresql queries
Add support for profiling Postgresql queries
5 votes -
4 votes
-
Monitor resources outside of the application (i.e. GDI object count)
I'll use GDI objects as they are what caused me to have the original idea, but it could apply to anything. I'm specifically thinking of performance counters as well.
So, you develop a winforms application. It uses graphics. One day your user complains it has bombed out. It turns out there's a GDI leak because a button has a graphic that isn't disposed. So I'd like to see GDI count mapped to execution of code.
Something like GDI could almost be recorded for each instruction, but performance counters could just be snapped each second or so allowing you to home…
3 votes -
Display timings on the timeline in absolute time
On the timeline, the timings are relative to the beginning of the profiling session - 1:00 elapsed, 2:00 elapsed, etc. Since my app generates logs with timestamps, it'd be useful to have the timings on the timeline relate to the system time at that point so that I could line them up with logs easily.
2 votes -
Provide easy filtering based on type/method
The profiler has a couple filtering options - "with source" and "complex stack traces" but both of those are far too coarse.
My suggestion is to provide a way to have a list of types/methods (perhaps with rx) that can be filtered out of a profile run from the UI.
Preferably with a choice where "filtering" means either "exclude timings entirely" (for completely ignoring items) or "roll up timings into caller" (for simplifying redundant/noisy methods).
Bonus: right-click on a method and have "ignore this type/namespace/method" kind of thing.
6 votes -
Add support for profiling Oracle SQL queries
It would be great to have a feature similar to profiling and tracing SQL Server's queries, but for Oracle also.
9 votes -
Integrate some form of code coverage detail.
Since you can track line level times, it should be possible to integrate a code coverage 'mode' into ANTS PP or create a tool that does this. VS2010 provides it, but only thru unit tests which is nice if you have all those unit tests, but often time you don't on legacy projects. It would be nice to be able to run my app under a code coverage tool, exercise its options, save the report, run again, merge reports and keep doing this until I exercise as much as I can.
3 votes -
Line level timings for Silverlight
Although work-aroundable (either by deduction, or moving suspect code into a separate method), it's pretty labour intensive.
12 votes -
A search box for namespaces to allow these to be filtered
Currently there is only a search box for Method Names and not namespaces.
3 votes -
Support for remote profiling
This is requested fairly frequently!
11 votes -
Ability to easily compare result sets
Currently you can only compare result sets using our ANTS XML tool. It would be nice if such a feature could be built into the profiler.
Developers could then show their boss/decision maker exactly how much ANTS has helped them to improve their application's performance, which I think could really help us to sell even more licenses.
12 votes -
Add ability to save results using Profiler Standard Edition
Why do I need the Professional edition to save results to share with other members of my team?
1 vote
- Don't see your idea?