Improve app performance with Android Studio Profilers (Google I/O '18)
Why Stack Overflow recommends it
Stack Overflow users link it when answering questions about profiling Android apps, such as finding CPU-heavy threads, tracking allocations and reading the Memory Profiler's instance counts.
Recommended by 2 people across 2 different months, 2019 to 2023.
What it is
A Google I/O 2018 session from the Android Developers channel showing how to diagnose app performance problems with the Android Studio Profilers. It walks through the CPU, memory and network profilers with worked examples and covers what was new at the time.
What people said
Recommends Morrison Chang · 2021-02 · Stack Overflow · on “Creating stream from canvas using custom capturer(webrtc)”
If you are able to capture data with Memory Profiler [LINK] , you should investigate what are those objects. Generally using `new` in a loop is a bad idea as allocations occur there (but not necessary GC). Also see: Improve app performance with Android Studio Profilers (Google I/O '18) [this video]
Recommends Francois Dermu · 2023-05 · Stack Overflow · on “How can I find out what thread is using CPU in my android application”
You can use the built-in profiler in Android Studio: [LINK] [LINK] You can also run adb shell top but this will only give you CPU and MEM usage per process not thread. For more info check out this Google I/O presentation: [this video]
Cites Axifive · 2019-01 · Stack Overflow · on “What is the "total count" metric in Android Studio's Memory Profiler?”
Yes this is the total number of instance, that has allocated before the selected period + memory allocation for the selected period (or as you says: number of objects(instances) in the heap that haven't been deallocated for the selected period) You can see this here [this video] On the Instance View, it can be seen that three int arrays were allocated at 1m, but not deallocated (for the selected period).