starpu-max/doc/doxygen/chapters/api/scheduling_context_hypervisor.doxy

/*
 * This file is part of the StarPU Handbook.
 * Copyright (C) 2009--2011  Universit@'e de Bordeaux 1
 * Copyright (C) 2010, 2011, 2012, 2013  Centre National de la Recherche Scientifique
 * Copyright (C) 2011, 2012 Institut National de Recherche en Informatique et Automatique
 * See the file version.doxy for copying conditions.
 */

/*! \defgroup API_Scheduling_Context_Hypervisor Scheduling Context Hypervisor

\struct sc_hypervisor_policy
\ingroup API_Scheduling_Context_Hypervisor
This structure contains all the methods that implement a hypervisor resizing policy.
\var sc_hypervisor_policy::name
        Indicates the name of the policy, if there is not a custom policy, the policy corresponding to this name will be used by the hypervisor
\var sc_hypervisor_policy::custom
        Indicates whether the policy is custom or not
\var sc_hypervisor_policy::handle_idle_cycle
        It is called whenever the indicated worker executes another idle cycle in sched_ctx
\var sc_hypervisor_policy::handle_pushed_task
        It is called whenever a task is pushed on the worker’s queue corresponding to the context sched_ctx
\var sc_hypervisor_policy::handle_poped_task
        It is called whenever a task is poped from the worker’s queue corresponding to the context sched_ctx
\var sc_hypervisor_policy::handle_idle_end
        It is called whenever a task is executed on the indicated worker and context after a long period of idle time
\var sc_hypervisor_policy::handle_post_exec_hook
        It is called whenever a tag task has just been executed. The table of resize requests is provided as well as the tag

\struct sc_hypervisor_policy_config
\ingroup API_Scheduling_Context_Hypervisor
This structure contains all configuration information of a
context. It contains configuration information for each context, which
can be used to construct new resize strategies.
\var sc_hypervisor_policy_config::min_nworkers
        Indicates the minimum number of workers needed by the context
\var sc_hypervisor_policy_config::max_nworkers
        Indicates the maximum number of workers needed by the context
\var sc_hypervisor_policy_config::granularity
        Indicates the workers granularity of the context
\var sc_hypervisor_policy_config::priority
        Indicates the priority of each worker in the context
\var sc_hypervisor_policy_config::max_idle
        Indicates the maximum idle time accepted before a resize is triggered
\var sc_hypervisor_policy_config::fixed_workers
        Indicates which workers can be moved and which ones are fixed
\var sc_hypervisor_policy_config:: new_workers_max_idle
        Indicates the maximum idle time accepted before a resize is triggered for the workers that just arrived in the new context

\struct sc_hypervisor_wrapper
\ingroup API_Scheduling_Context_Hypervisor
This structure is a wrapper of the contexts available in StarPU
and contains all information about a context obtained by incrementing
the performance counters.
\var sc_hypervisor_wrapper::sched_ctx
        The context wrapped
\var sc_hypervisor_wrapper::config
        The corresponding resize configuration
\var sc_hypervisor_wrapper::current_idle_time
        The idle time counter of each worker of the context
\var sc_hypervisor_wrapper::pushed_tasks
        The number of pushed tasks of each worker of the context
\var sc_hypervisor_wrapper::poped_tasks
        The number of poped tasks of each worker of the context
\var sc_hypervisor_wrapper::total_flops
        The total number of flops to execute by the context
\var sc_hypervisor_wrapper::total_elapsed_flops
        The number of flops executed by each workers of the context
\var sc_hypervisor_wrapper::elapsed_flops
        The number of flops executed by each worker of the context from last resize
\var sc_hypervisor_wrapper::remaining_flops
        The number of flops that still have to be executed by the workers in the context
\var sc_hypervisor_wrapper::start_time
        The time when he started executed
\var sc_hypervisor_wrapper::resize_ack
        The structure confirming the last resize finished and a new one can be done

\struct sc_hypervisor_resize_ack
\ingroup API_Scheduling_Context_Hypervisor
This structures checks if the workers moved to another context
are actually taken into account in that context.
\var sc_hypervisor_resize_ack::receiver_sched_ctx
        The context receiving the new workers
\var sc_hypervisor_resize_ack::moved_workers
        The workers moved to the receiver context
\var sc_hypervisor_resize_ack::nmoved_workers
        The number of workers moved
\var sc_hypervisor_resize_ack::acked_workers
        If the value corresponding to a worker is 1, this one is taken
	into account in the new context if 0 not yet

\struct sc_hypervisor_policy_task_pool
task wrapper linked list
\ingroup API_Scheduling_Context_Hypervisor
\var sc_hypervisor_policy_task_pool::cl
Which codelet has been executed
\var sc_hypervisor_policy_task_pool::footprint
Task footprint key
\var sc_hypervisor_policy_task_pool::sched_ctx_id
Context the task belongs to
\var sc_hypervisor_policy_task_pool::n
Number of tasks of this kind
\var sc_hypervisor_policy_task_pool::next
Other task kinds

@name Managing the hypervisor
\ingroup API_Scheduling_Context_Hypervisor

There is a single hypervisor that is in charge of resizing contexts
and the resizing strategy is chosen at the initialization of the
hypervisor. A single resize can be done at a time.

The Scheduling Context Hypervisor Plugin provides a series of
performance counters to StarPU. By incrementing them, StarPU can help
the hypervisor in the resizing decision making process. TODO maybe
they should be hidden to the user

\fn struct starpu_sched_ctx_performance_counters *sc_hypervisor_init(struct sc_hypervisor_policy * policy)
\ingroup API_Scheduling_Context_Hypervisor
Initializes the hypervisor to use the strategy provided as parameter
and creates the performance counters (see \ref Performance_Counters).
These performance counters represent actually some callbacks that will
be used by the contexts to notify the information needed by the
hypervisor.

Note: The Hypervisor is actually a worker that takes this role once
certain conditions trigger the resizing process (there is no
additional thread assigned to the hypervisor).

\fn void sc_hypervisor_shutdown(void)
\ingroup API_Scheduling_Context_Hypervisor
The hypervisor and all information concerning it is cleaned. There is
no synchronization between this function and starpu_shutdown(). Thus,
this should be called after starpu_shutdown(), because the performance
counters will still need allocated callback functions.

@name Registering Scheduling Contexts to the hypervisor
\ingroup API_Scheduling_Context_Hypervisor

Scheduling Contexts that have to be resized by the hypervisor must be
first registered to the hypervisor. Whenever we want to exclude
contexts from the resizing process we have to unregister them from the
hypervisor.

\fn void sc_hypervisor_register_ctx(unsigned sched_ctx, double total_flops)
\ingroup API_Scheduling_Context_Hypervisor
Register the context to the hypervisor, and indicate the number of
flops the context will execute (needed for Gflops rate based strategy
see \ref Resizing_strategies or any other custom strategy needing it, for
the others we can pass 0.0)

\fn void sc_hypervisor_unregister_ctx (unsigned sched_ctx)
\ingroup API_Scheduling_Context_Hypervisor
Unregister the context from the hypervisor.

@name The user’s input in the resizing process
\ingroup API_Scheduling_Context_Hypervisor

The user can totally forbid the resizing of a certain context or can
then change his mind and allow it (in this case the resizing is
managed by the hypervisor, that can forbid it or allow it)

\fn void sc_hypervisor_stop_resize(unsigned sched_ctx)
\ingroup API_Scheduling_Context_Hypervisor
Forbid resizing of a context

\fn void sc_hypervisor_start_resize(unsigned sched_ctx)
\ingroup API_Scheduling_Context_Hypervisor
Allow resizing of a context. The user can then provide information to
the hypervisor concerning the conditions of resizing.

\fn void sc_hypervisor_ioctl(unsigned sched_ctx, ...)
\ingroup API_Scheduling_Context_Hypervisor
Inputs conditions to the context sched_ctx with the following
arguments. The argument list must be zero-terminated.

\def HYPERVISOR_MAX_IDLE
\ingroup API_Scheduling_Context_Hypervisor
This macro is used when calling sc_hypervisor_ioctl() and must be
followed by 3 arguments: an array of int for the workerids to apply
the condition, an int to indicate the size of the array, and a double
value indicating the maximum idle time allowed for a worker before the
resizing process should be triggered

\def HYPERVISOR_PRIORITY
\ingroup API_Scheduling_Context_Hypervisor
This macro is used when calling sc_hypervisor_ioctl() and must be
followed by 3 arguments: an array of int for the workerids to apply
the condition, an int to indicate the size of the array, and an int
value indicating the priority of the workers previously mentioned. The
workers with the smallest priority are moved the first.

\def HYPERVISOR_MIN_WORKERS
\ingroup API_Scheduling_Context_Hypervisor
This macro is used when calling sc_hypervisor_ioctl() and must be
followed by 1 argument(int) indicating the minimum number of workers a
context should have, underneath this limit the context cannot execute.

\def HYPERVISOR_MAX_WORKERS
\ingroup API_Scheduling_Context_Hypervisor
This macro is used when calling sc_hypervisor_ioctl() and must be
followed by 1 argument(int) indicating the maximum number of workers a
context should have, above this limit the context would not be able to
scale

\def HYPERVISOR_GRANULARITY
\ingroup API_Scheduling_Context_Hypervisor
This macro is used when calling sc_hypervisor_ioctl() and must be
followed by 1 argument(int) indicating the granularity of the resizing
process (the number of workers should be moved from the context once
it is resized) This parameter is ignore for the Gflops rate based
strategy see Resizing strategies, the number of workers that have to
be moved is calculated by the strategy.

\def HYPERVISOR_FIXED_WORKERS
\ingroup API_Scheduling_Context_Hypervisor
This macro is used when calling sc_hypervisor_ioctl() and must be
followed by 2 arguments: an array of int for the workerids to apply
the condition and an int to indicate the size of the array. These
workers are not allowed to be moved from the context.

\def HYPERVISOR_MIN_TASKS
\ingroup API_Scheduling_Context_Hypervisor
This macro is used when calling sc_hypervisor_ioctl() and must be
followed by 1 argument (int) that indicated the minimum number of
tasks that have to be executed before the context could be resized.
This parameter is ignored for the Application Driven strategy see
Resizing strategies where the user indicates exactly when the resize
should be done.

\def HYPERVISOR_NEW_WORKERS_MAX_IDLE
\ingroup API_Scheduling_Context_Hypervisor
This macro is used when calling sc_hypervisor_ioctl() and must be
followed by 1 argument, a double value indicating the maximum idle
time allowed for workers that have just been moved from other contexts
in the current context.

\def HYPERVISOR_TIME_TO_APPLY
\ingroup API_Scheduling_Context_Hypervisor
This macro is used when calling sc_hypervisor_ioctl() and must be
followed by 1 argument (int) indicating the tag an executed task
should have such that this configuration should be taken into account.

@name Defining a new hypervisor policy
\ingroup API_Scheduling_Context_Hypervisor

While Scheduling Context Hypervisor Plugin comes with a variety of
resizing policies (see \ref Resizing_strategies), it may sometimes be
desirable to implement custom policies to address specific problems.
The API described below allows users to write their own resizing policy.

Here an example of how to define a new policy

\code{.c}
struct sc_hypervisor_policy dummy_policy =
{
       .handle_poped_task = dummy_handle_poped_task,
       .handle_pushed_task = dummy_handle_pushed_task,
       .handle_idle_cycle = dummy_handle_idle_cycle,
       .handle_idle_end = dummy_handle_idle_end,
       .handle_post_exec_hook = dummy_handle_post_exec_hook,
       .custom = 1,
       .name = "dummy"
};
\endcode

\fn void sc_hypervisor_move_workers(unsigned sender_sched_ctx, unsigned receiver_sched_ctx, int *workers_to_move, unsigned nworkers_to_move, unsigned now);
\ingroup API_Scheduling_Context_Hypervisor
    Moves workers from one context to another

\fn struct sc_hypervisor_policy_config * sc_hypervisor_get_config(unsigned sched_ctx);
\ingroup API_Scheduling_Context_Hypervisor
    Returns the configuration structure of a context

\fn int * sc_hypervisor_get_sched_ctxs();
\ingroup API_Scheduling_Context_Hypervisor
    Gets the contexts managed by the hypervisor

\fn int sc_hypervisor_get_nsched_ctxs();
\ingroup API_Scheduling_Context_Hypervisor
    Gets the number of contexts managed by the hypervisor

\fn struct sc_hypervisor_wrapper * sc_hypervisor_get_wrapper(unsigned sched_ctx);
\ingroup API_Scheduling_Context_Hypervisor
    Returns the wrapper corresponding the context \p sched_ctx

\fn double sc_hypervisor_get_elapsed_flops_per_sched_ctx(struct sc_hypervisor_wrapper * sc_w);
\ingroup API_Scheduling_Context_Hypervisor
    Returns the flops of a context elapsed from the last resize

\fn char * sc_hypervisor_get_policy();
\ingroup API_Scheduling_Context_Hypervisor
    Returns the name of the resizing policy the hypervisor uses

*/