137. MPI's Support for Libraries
Up: Introduction
Next: Basic Concepts
Previous: Features Needed to Support Libraries
The corresponding
concepts that MPI provides, specifically to support robust libraries, are
as follows:
- Contexts of communication,
- Groups of processes,
- Virtual topologies,
- Attribute caching,
- Communicators.
Communicators (see [21,53,57]) encapsulate all of
these ideas in order to provide the appropriate scope for all communication
operations in MPI. Communicators are divided into two kinds:
intra-communicators for operations within a single group of processes and
inter-communicators for operations between two groups of
processes.
Caching. Communicators (see
below) provide a ``caching'' mechanism that allows one to
associate new attributes with communicators, on
par with MPI built-in
features. This can be used by advanced users to adorn communicators further,
and by MPI to implement some communicator functions. For example, the
virtual-topology functions described in
Chapter Process Topologies
are likely to be supported this way.
Groups. Groups
define an ordered collection of processes, each with a rank, and it is this
group that defines the low-level names for inter-process communication (ranks
are used for sending and receiving). Thus, groups define a scope for process
names in point-to-point communication. In addition, groups define the scope
of collective operations. Groups may be manipulated separately from
communicators in MPI, but only communicators can be used in
communication operations.
Intra-communicators. The most commonly used means for message
passing in MPI is via intra-communicators. Intra-communicators contain an
instance of a group, contexts of communication for both point-to-point and
collective communication, and the ability to include virtual topology and
other attributes.
These features work as follows:
- Contexts provide the ability to have separate safe ``universes''
of message-passing in MPI. A context is akin to an additional
tag that differentiates messages.
The system manages this differentiation process.
The use of separate communication
contexts by distinct libraries (or distinct library invocations)
insulates communication internal to the library execution from
external communication. This allows the invocation of the library even if
there are pending communications
on ``other'' communicators, and avoids the need to
synchronize entry or exit into library code. Pending point-to-point
communications are also guaranteed not to interfere with collective
communications within a single communicator.
- Groups define the participants in the communication (see above)
of a communicator.
- A virtual topology defines a special mapping of the ranks in a
group to and from a topology. Special constructors for
communicators are defined in Chapter Process Topologies
to provide
this feature. Intra-communicators as described in this chapter do
not have topologies.
- Attributes define the local information that the user or
library has added to a communicator for later reference.
Advice to users.
The practice in many communication libraries is that there is
a unique, predefined communication universe that includes all
processes available when the parallel program is initiated; the processes are
assigned consecutive ranks. Participants in a point-to-point
communication are identified by their rank; a collective communication
(such as broadcast) always involves all processes. This practice can be
followed in MPI by using the predefined communicator
MPI_COMM_WORLD. Users who are satisfied with this practice
can plug in MPI_COMM_WORLD wherever a communicator argument
is required, and can consequently disregard the rest of this chapter.
( End of advice to users.)
Inter-communicators. The discussion has dealt so far with intra-communication:
communication
within a group. MPI also supports inter-communication:
communication
between two non-overlapping groups. When an application is built by composing
several parallel modules, it is convenient to allow one module to communicate
with another using local ranks for addressing within the second module. This
is especially convenient in a client-server computing paradigm, where either
client or server are parallel. The support of inter-communication
also provides a mechanism for the extension of MPI to a dynamic model where
not all processes are preallocated at initialization time. In such a
situation, it becomes necessary to support communication across ``universes.''
Inter-communication is supported by objects called
inter-communicators.
These objects bind two groups together with communication contexts shared by
both groups.
For inter-communicators, these features work as follows:
- Contexts provide the ability to have
a separate safe ``universe''
of message-passing between the two groups. A send in the local
group is always a receive in the remote group, and vice versa.
The system manages this differentiation process.
The use of separate communication
contexts by distinct libraries (or distinct library invocations)
insulates communication internal to the library execution from
external communication. This allows the invocation of the library even if
there are pending communications
on ``other'' communicators, and avoids the need to
synchronize entry or exit into library
code.
- A local and remote group specify the recipients and destinations
for an inter-communicator.
- Virtual topology is undefined for an inter-communicator.
- As before,
attributes cache defines the local information that the user or
library has added to a communicator for later reference.
MPI provides mechanisms for creating and manipulating inter-communicators.
They are used for point-to-point
and collective
communication in an related manner to
intra-communicators. Users who do not need inter-communication
in their applications can safely ignore this extension.
Users
who require inter-communication between overlapping groups
must layer
this capability on top of MPI.
Up: Introduction
Next: Basic Concepts
Previous: Features Needed to Support Libraries
Return to MPI-3.1 Standard Index
Return to MPI Forum Home Page
(Unofficial) MPI-3.1 of June 4, 2015
HTML Generated on June 4, 2015