Hot answers tagged profiling
GNU time can report a bit more information than the version built into Bash; use command time rather than just time to invoke it, and see the man page or info for details.
zsh has a more powerful built-in time command than bash has, and the zsh version can report memory statistics. Even if you don't regularly use zsh as your day-to-day shell, you can just run it when you need to gather these kinds of statistics. Set the TIMEFMT environment variable to indicate the output you want. Here is what I have in my .zshrc file ...
Windows 7 does indeed cache all kinds of files, including applications, in memory. Your guess is probably correct, the fact that the processes are in memory makes them start much more quickly. Not only are the executable files themselves cached, but the DLLs they require are also loaded and ready.
There is no way currently for Firefox to profile it's extensions without a significant overhead due to the way the Extensions API is built. This extension will tell you a bit more about your startup times: About Startup And Mozilla conducts regular tests on the addons every week. Here's the listing of the slowest addons (updated weekly) And if you want ...
What you want is cacti.
Use ps, time, and top.
iotop and dstat are very very handy tools for identifying performance issues. I recommend specifically 'dstat -af' to find out what resource is (at any given time) the limiting factor.
That's just what random-pausing (stackshots) is for. It's easy to do in python using ctrl-C. Here's a brief explanation. If you want visibility into the C code, an alternative is Zoom.
I'd use XBench or GeekBench to determine where your hardware bottlenecks are, but you'll likely learn what you already know: HDDs are slow and your processor is old. As for an extended profile of your usage habits, I'm unsure of the benefit that you're expecting to derive. All well-written Mac apps will not leak memory and will idle when not in use, so the ...
If you aren't quite happy with the way RRDT-based tools work, you may want to try Zabbix. CPU, memory, disk and network are in the default template for all OSes, so it should be relatively configuration free. You can get Zabbix packages (both server and client) with many Linux distributions' package management system directly. It uses more traditional ...
You can use cachegrind. Memory bandwidth is essentially the same as misses in the lowest-level cache.
GKrellm. GKrellM is a single process stack of system monitors which supports applying themes to match its appearance to your window manager, Gtk, or any other theme.
conhost.exe is the console-hosting process, that is launched by the Command Prompt (cmd). It is present because you are using a command-line program. src2srcml.exe is part of the srcML toolkit, and is present probably because you are manipulating Source Markup or XML files. As to why it is slower on the first call, as was already noted by MoJo, it is ...
The kernel caches I/O read from incoming block devices in free RAM. Unless your file is very large you are probably hitting the cached data in RAM and not actually causing any I/O. A very quick search and reading this Stack Overflow answer reveals that this might do it: sync && echo 1 > /proc/sys/vm/drop_caches so I would try issuing those ...
Only top voted, non community-wiki answers of a minimum length are eligible