wiki:AppPlan

Version 25 (modified by davea, 13 years ago) (diff)

--

Application planning

Application planning is a mechanism that lets the scheduler decide, using project-supplied logic:

  • whether an application should run on a particular host;
  • what resources it will use;
  • how fast it is expected to run.

It works as follows.

An app version has an associated plan_class: a character string, possibly empty. See how to specify an app version's plan class.

The scheduler is linked with a function

bool app_plan(SCHEDULER_REQUEST &sreq, char* plan_class, HOST_USAGE&);

The sreq argument describes the host. It contains:

  • in sreq.host field, a description of the host's hardware, including:
    • In p_vendor and p_model, the processor type
    • In p_features, the processor features (e.g., fpu tsc pae nx sse sse2 mmx)
    • In m_nbytes, the amount of RAM
  • in sreq.coprocs, a list of the hosts's coprocessors.
  • in core_client_version, the client's version number in MMmmRR form.

When called with a particular SCHEDULER_REQUEST and plan class, the function returns true if the host's resources are sufficient for apps of that class. If true, it populates the HOST_USAGE structure:

struct HOST_USAGE {
   double ncudas;     // number of NVIDIA GPUs used
   double natis;      // number of ATI GPUs used
   double gpu_ram;    // max amount of GPU RAM used
   double avg_ncpus;  // avg #CPUs used by app (may be fractional)
   double max_ncpus;  // max #CPUs used (not currently used for anything)
   double projected_flops;
      // an estimate of the actual FLOPS.
      // used to select versions, so make it higher for the preferred version
   double peak_flops;
      // the peak FLOPS of the devices to be used
   char cmdline[256]; // passed to the app as a cmdline argument;
                      // this can be used, e.g. to control the # of threads used
};

When deciding whether to send a job to a host, the scheduler examines all latest-version app_versions for the platform, calls app_plan() for each, and selects the one for which projected_flops is greatest.

If gpu_ram is nonzero, the BOINC client (6.10.25+) won't start the app unless that much RAM is available on the allocated GPU.

You are free to define your own set of plan classes, and to link your own app_plan() function with the scheduler. The BOINC scheduler comes with a default app_plan() (in sched/sched_customize.cpp). This defines the following plan classes:

mt
An application that can use anywhere from 1 to 64 threads, and whose speedup with N CPUs is .95N. It is passed a command-line argument --nthreads N.
nci
A non-CPU-intensive application that uses 1% of a CPU (this will cause the BOINC client 6.7+ to run it at non-idle priority).
sse3
A CPU app that requires the SSE3 CPU feature.
vbox32
An app that runs in a 32-bit VirtualBox VM
vbox64
An app that runs in a 64-bit VirtualBox VM

... and a number of GPU plan classes.

Note: plan classes containing the substrings 'nvidia', 'cuda', and 'ati' are reserved for GPU applications.