Continuous profiling for analysis of CPU and memory usage, down to the line number and throughout time. Saving infrastructure cost, improving performance, and increasing reliability.
It would be interesting to enhance parca with data for I/O operations per second and latency in addition to the CPU-based profiling, so a user can figure out if a process is CPU- or IO-bound.
For disks, I guess right now it's possible to take a look at the vfs_read syscalls, it could be helpful to visualize I/O access in a separate view.
brancz
commented
10 months ago
It would be interesting to enhance parca with data for I/O operations per second and latency in addition to the CPU-based profiling, so a user can figure out if a process is CPU- or IO-bound. For disks, I guess right now it's possible to take a look at the vfs_read syscalls, it could be helpful to visualize I/O access in a separate view.
https://github.com/iovisor/bpftrace/blob/master/docs/tutorial_one_liners.md#lesson-7-timing-reads https://github.com/iovisor/bcc/blob/master/tools/biolatency_example.txt https://github.com/iovisor/bcc/blob/master/tools/biotop_example.txt