Linux – cpu usage per process

What you want to get its the instant CPU usage (kind of)...

Actually, the instant CPU usage for a process does not exists. Instead you have to make two measurements and calculate the average CPU usage, the formula is quite simple:

AvgCpuUsed = [TotalCPUTime(process,time2) - TotalCPUTime(process,time1)] / [time2-time1]

The lower Time2 and Time1 difference is, the more "instant" your measurement will be. Windows Task Manager calculate the CPU use with an interval of one second. I've found that is more than enough and you might even consider doing it in 5 seconds intervals cause the act of measuring itself takes up CPU cycles...

So, first, to get the average CPU time

    using System.Diagnostics;

float GetAverageCPULoad(int procID, DateTme from, DateTime, to)
{
  // For the current process
  //Process proc = Process.GetCurrentProcess();
  // Or for any other process given its id
  Process proc = Process.GetProcessById(procID);
  System.TimeSpan lifeInterval = (to - from);
  // Get the CPU use
  float CPULoad = (proc.TotalProcessorTime.TotalMilliseconds / lifeInterval.TotalMilliseconds) * 100;
  // You need to take the number of present cores into account
  return CPULoad / System.Environment.ProcessorCount;
}

now, for the "instant" CPU load you'll need an specialized class:

 class ProcLoad
{
  // Last time you checked for a process
  public Dictionary<int, DateTime> lastCheckedDict = new Dictionary<int, DateTime>();

  public float GetCPULoad(int procID)
  {
    if (lastCheckedDict.ContainsKey(procID))
    {
      DateTime last = lastCheckedDict[procID];
      lastCheckedDict[procID] = DateTime.Now;
      return GetAverageCPULoad(procID, last, lastCheckedDict[procID]);
    }
    else
    {
      lastCheckedDict.Add(procID, DateTime.Now);
      return 0;
    }
  }
}

You should call that class from a timer (or whatever interval method you like) for each process you want to monitor, if you want all the processes just use the Process.GetProcesses static method

With ps or similar tools you will only get the amount of memory pages allocated by that process. This number is correct, but:

• does not reflect the actual amount of memory used by the application, only the amount of memory reserved for it

• can be misleading if pages are shared, for example by several threads or by using dynamically linked libraries

If you really want to know what amount of memory your application actually uses, you need to run it within a profiler. For example, Valgrind can give you insights about the amount of memory used, and, more importantly, about possible memory leaks in your program. The heap profiler tool of Valgrind is called 'massif':

Massif is a heap profiler. It performs detailed heap profiling by taking regular snapshots of a program's heap. It produces a graph showing heap usage over time, including information about which parts of the program are responsible for the most memory allocations. The graph is supplemented by a text or HTML file that includes more information for determining where the most memory is being allocated. Massif runs programs about 20x slower than normal.

As explained in the Valgrind documentation, you need to run the program through Valgrind:

valgrind --tool=massif <executable> <arguments>

Massif writes a dump of memory usage snapshots (e.g. massif.out.12345). These provide, (1) a timeline of memory usage, (2) for each snapshot, a record of where in your program memory was allocated. A great graphical tool for analyzing these files is massif-visualizer. But I found ms_print, a simple text-based tool shipped with Valgrind, to be of great help already.

To find memory leaks, use the (default) memcheck tool of valgrind.