Class mpi.Intracomm

java.lang.Object
    |
    +----mpi.Comm
            |
            +----mpi.Intracomm

Subclasses:

Cartcomm, Graphcomm

public class Intracomm

extends Comm

Fields inherited from class mpi.Comm

handle, NULL, SELF, WORLD

 

Constructor Summary

Intracomm()

Intracomm(int Type)

Intracomm(long _handle)

 

Method Summary
void	Allgather(java.lang.Object sendbuf, int sendoffset, int sendcount, Datatype sendtype, java.lang.Object recvbuf, int recvoffset, int recvcount, Datatype recvtype) Similar to Gather, but all processes receive the result.
void	Allgatherv(java.lang.Object sendbuf, int sendoffset, int sendcount, Datatype sendtype, java.lang.Object recvbuf, int recvoffset, int[] recvcounts, int[] displs, Datatype recvtype) Similar to Gatherv, but all processes receive the result.
void	Allreduce(java.lang.Object sendbuf, int sendoffset, java.lang.Object recvbuf, int recvoffset, int count, Datatype datatype, Op op) Same as reduce except that the result appears in receive buffer of all process in the group.
void	Alltoall(java.lang.Object sendbuf, int sendoffset, int sendcount, Datatype sendtype, java.lang.Object recvbuf, int recvoffset, int recvcount, Datatype recvtype) Extension of Allgather to the case where each process sends distinct data to each of the receivers.
void	Alltoallv(java.lang.Object sendbuf, int sendoffset, int[] sendcounts, int[] sdispls, Datatype sendtype, java.lang.Object recvbuf, int recvoffset, int[] recvcounts, int[] displs, Datatype recvtype) Adds flexibility to Alltoall: location of data for send is specified by sdispls and location to place data on receive side is specified by rdispls.
void	Barrier() A call to Barrier blocks the caller until all process in the group have called it.
void	Bcast(java.lang.Object buf, int offset, int count, Datatype datatype, int root) Broadcast a message from the process with rank root to all processes of the group.
java.lang.Object	clone()
Intracomm	Creat(Group group) Create a new communicator.
Cartcomm	Create_cart(int[] dims, boolean[] periods, boolean reorder) Create a Cartesian topology communicator whose group is a subset of the group of this communicator.
Graphcomm	Create_graph(int[] index, int[] edges, boolean reorder) Create a graph topology communicator whose group is a subset of the group of this communicator.
void	Gather(java.lang.Object sendbuf, int sendoffset, int sendcount, Datatype sendtype, java.lang.Object recvbuf, int recvoffset, int recvcount, Datatype recvtype, int root) Each process sends the contents of its send buffer to the root process.
void	Gatherv(java.lang.Object sendbuf, int sendoffset, int sendcount, Datatype sendtype, java.lang.Object recvbuf, int recvoffset, int[] recvcounts, int[] displs, Datatype recvtype, int root) Extends functionality of Gather by allowing varying counts of data from each process.
long	GetGraph(int[] index, int[] edges, boolean reorder)
void	Reduce(java.lang.Object sendbuf, int sendoffset, java.lang.Object recvbuf, int recvoffset, int count, Datatype datatype, Op op, int root) Combine elements in input buffer of each process using the reduce operation, and return the combined value in the output buffer of the root process.
void	Reduce_scatter(java.lang.Object sendbuf, int sendoffset, java.lang.Object recvbuf, int recvoffset, int[] recvcounts, Datatype datatype, Op op) Combine elements in input buffer of each process using the reduce operation, and scatter the combined values over the output buffers of the processes.
void	Scan(java.lang.Object sendbuf, int sendoffset, java.lang.Object recvbuf, int recvoffset, int count, Datatype datatype, Op op) Perform a prefix reduction on data distributed across the group.
void	Scatter(java.lang.Object sendbuf, int sendoffset, int sendcount, Datatype sendtype, java.lang.Object recvbuf, int recvoffset, int recvcount, Datatype recvtype, int root) Inverse of the operation Gather.
void	Scatterv(java.lang.Object sendbuf, int sendoffset, int[] sendcounts, int[] displs, Datatype sendtype, java.lang.Object recvbuf, int recvoffset, int recvcount, Datatype recvtype, int root) Inverse of the operation Gatherv.
Intracomm	Split(int colour, int key) Partition the group associated with this communicator and create a new communicator within each subgroup.

 

Methods inherited from class mpi.Comm

Abort, Attr_delete, Attr_get, Attr_put, Bsend, Bsend_init, clone, Compare, Create_intercomm, dup, Errhandler_set, Errorhandler_get, Free, GetComm, GetIntercomm, Group, Ibsend, Iprobe, Irecv, Irsend, Isend, Issend, Pack, Pack_size, Probe, Rank, Recv, Recv_init, Rsend, Rsend_init, Send, Send_init, Sendrecv, Sendrecv_replace, Size, Ssend, Ssend_init, Test_inter, Topo_test, Unpack

 

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

 

Constructor Detail

Intracomm

public Intracomm()

Intracomm

public Intracomm(int Type)

Intracomm

public Intracomm(long _handle)

clone

public java.lang.Object clone()

Overrides:

clone in class Comm

Split

public Intracomm Split(int colour,
                       int key)

Partition the group associated with this communicator and create a new communicator within each subgroup.

color	control of subset assignment
key	control of rank assignment
returns:	new communicator

Java binding of the MPI operation MPI_COMM_SPLIT.

Creat

public Intracomm Creat(Group group)

Create a new communicator.

group	group which is a subset of the group of this communicator
returns:	new communicator

Java binding of the MPI operation MPI_COMM_CREATE.

Barrier

public void Barrier()

A call to Barrier blocks the caller until all process in the group have called it.

Java binding of the MPI operation MPI_BARRIER.

Bcast

public void Bcast(java.lang.Object buf,
                  int offset,
                  int count,
                  Datatype datatype,
                  int root)

Broadcast a message from the process with rank root to all processes of the group.

buf	buffer array
offset	initial offset in buffer
count	number of items in buffer
datatype	datatype of each item in buffer
root	rank of broadcast root

Java binding of the MPI operation MPI_BCST.

Gather

public void Gather(java.lang.Object sendbuf,
                   int sendoffset,
                   int sendcount,
                   Datatype sendtype,
                   java.lang.Object recvbuf,
                   int recvoffset,
                   int recvcount,
                   Datatype recvtype,
                   int root)

Each process sends the contents of its send buffer to the root process.

sendbuf	send buffer array
sendoffset	initial offset in send buffer
sendcount	number of items to send
sendtype	datatype of each item in send buffer
recvbuf	receive buffer array
recvoffset	initial offset in receive buffer
recvcount	number of items to receive
recvtype	datatype of each item in receive buffer
root	rank of receiving process

Java binding of the MPI operation MPI_GATHER.

Gatherv

public void Gatherv(java.lang.Object sendbuf,
                    int sendoffset,
                    int sendcount,
                    Datatype sendtype,
                    java.lang.Object recvbuf,
                    int recvoffset,
                    int[] recvcounts,
                    int[] displs,
                    Datatype recvtype,
                    int root)

Extends functionality of Gather by allowing varying counts of data from each process.

sendbuf	send buffer array
sendoffset	initial offset in send buffer
sendcount	number of items to send
sendtype	datatype of each item in send buffer
recvbuf	receive buffer array
recvoffset	initial offset in receive buffer
recvcounts	number of elements received from each process
displs	displacements at which to place incoming data
recvtype	datatype of each item in receive buffer
root	rank of receiving process

Java binding of the MPI operation MPI_GATHERV.

The sizes of arrays recvcounts and displs should be the size of the group. Entry i of displs specifies the displacement relative to element recvoffset of recvbuf at which to place incoming data.

Scatter

public void Scatter(java.lang.Object sendbuf,
                    int sendoffset,
                    int sendcount,
                    Datatype sendtype,
                    java.lang.Object recvbuf,
                    int recvoffset,
                    int recvcount,
                    Datatype recvtype,
                    int root)

Inverse of the operation Gather.

sendbuf	send buffer array
sendoffset	initial offset in send buffer
sendcount	number of items to send
sendtype	datatype of each item in send buffer
recvbuf	receive buffer array
recvoffset	initial offset in receive buffer
recvcount	number of items to receive
recvtype	datatype of each item in receive buffer
root	rank of sending process

Java binding of the MPI operation MPI_SCATTER.

Scatterv

public void Scatterv(java.lang.Object sendbuf,
                     int sendoffset,
                     int[] sendcounts,
                     int[] displs,
                     Datatype sendtype,
                     java.lang.Object recvbuf,
                     int recvoffset,
                     int recvcount,
                     Datatype recvtype,
                     int root)

Inverse of the operation Gatherv.

sendbuf	send buffer array
sendoffset	initial offset in send buffer
sendcounts	number of items sent to each process
displs	displacements from which to take outgoing data
sendtype	datatype of each item in send buffer
recvbuf	receive buffer array
recvoffset	initial offset in receive buffer
recvcount	number of items to receive
recvtype	datatype of each item in receive buffer
root	rank of sending process

Java binding of the MPI operation MPI_SCATTERV.

Allgather

public void Allgather(java.lang.Object sendbuf,
                      int sendoffset,
                      int sendcount,
                      Datatype sendtype,
                      java.lang.Object recvbuf,
                      int recvoffset,
                      int recvcount,
                      Datatype recvtype)

Similar to Gather, but all processes receive the result.

sendbuf	send buffer array
sendoffset	initial offset in send buffer
sendcount	number of items to send
sendtype	datatype of each item in send buffer
recvbuf	receive buffer array
recvoffset	initial offset in receive buffer
recvcount	number of items to receive
recvtype	datatype of each item in receive buffer

Java binding of the MPI operation MPI_ALLGATHER.

Allgatherv

public void Allgatherv(java.lang.Object sendbuf,
                       int sendoffset,
                       int sendcount,
                       Datatype sendtype,
                       java.lang.Object recvbuf,
                       int recvoffset,
                       int[] recvcounts,
                       int[] displs,
                       Datatype recvtype)

Similar to Gatherv, but all processes receive the result.

sendbuf	send buffer array
sendoffset	initial offset in send buffer
sendcount	number of items to send
sendtype	datatype of each item in send buffer
recvbuf	receive buffer array
recvoffset	initial offset in receive buffer
recvcounts	number of elements received from each process
displs	displacements at which to place incoming data
recvtype	datatype of each item in receive buffer

Java binding of the MPI operation MPI_ALLGATHERV.

Alltoall

public void Alltoall(java.lang.Object sendbuf,
                     int sendoffset,
                     int sendcount,
                     Datatype sendtype,
                     java.lang.Object recvbuf,
                     int recvoffset,
                     int recvcount,
                     Datatype recvtype)

Extension of Allgather to the case where each process sends distinct data to each of the receivers.

sendbuf	send buffer array
sendoffset	initial offset in send buffer
sendcount	number of items sent to each process
sendtype	datatype send buffer items
recvbuf	receive buffer array
recvoffset	initial offset in receive buffer
recvcount	number of items received from any process
recvtype	datatype of receive buffer items

Java binding of the MPI operation MPI_ALLTOALL.

Alltoallv

public void Alltoallv(java.lang.Object sendbuf,
                      int sendoffset,
                      int[] sendcounts,
                      int[] sdispls,
                      Datatype sendtype,
                      java.lang.Object recvbuf,
                      int recvoffset,
                      int[] recvcounts,
                      int[] displs,
                      Datatype recvtype)

Adds flexibility to Alltoall: location of data for send is specified by sdispls and location to place data on receive side is specified by rdispls.

sendbuf	send buffer array
sendoffset	initial offset in send buffer
sendcounts	number of items sent to each process
sdispls	displacements from which to take outgoing data
sendtype	datatype send buffer items
recvbuf	receive buffer array
recvoffset	initial offset in receive buffer
recvcounts	number of elements received from each process
rdispls	displacements at which to place incoming data
recvtype	datatype of each item in receive buffer

Java binding of the MPI operation MPI_ALLTOALLV.

Reduce

public void Reduce(java.lang.Object sendbuf,
                   int sendoffset,
                   java.lang.Object recvbuf,
                   int recvoffset,
                   int count,
                   Datatype datatype,
                   Op op,
                   int root)

Combine elements in input buffer of each process using the reduce operation, and return the combined value in the output buffer of the root process.

sendbuf	send buffer array
sendoffset	initial offset in send buffer
recvbuf	receive buffer array
recvoffset	initial offset in receive buffer
count	number of items in send buffer
datatype	data type of each item in send buffer
op	reduce operation
root	rank of root process

Java binding of the MPI operation MPI_REDUCE.

The predefined operations are available in Java as MPI.MAX, MPI.MIN, MPI.SUM, MPI.PROD, MPI.LAND, MPI.BAND, MPI.LOR, MPI.BOR, MPI.LXOR, MPI.BXOR, MPI.MINLOC and MPI.MAXLOC.

Allreduce

public void Allreduce(java.lang.Object sendbuf,
                      int sendoffset,
                      java.lang.Object recvbuf,
                      int recvoffset,
                      int count,
                      Datatype datatype,
                      Op op)

Same as reduce except that the result appears in receive buffer of all process in the group.

sendbuf	send buffer array
sendoffset	initial offset in send buffer
recvbuf	receive buffer array
recvoffset	initial offset in receive buffer
count	number of items in send buffer
datatype	data type of each item in send buffer
op	reduce operation

Java binding of the MPI operation MPI_ALLREDUCE.

Reduce_scatter

public void Reduce_scatter(java.lang.Object sendbuf,
                           int sendoffset,
                           java.lang.Object recvbuf,
                           int recvoffset,
                           int[] recvcounts,
                           Datatype datatype,
                           Op op)

Combine elements in input buffer of each process using the reduce operation, and scatter the combined values over the output buffers of the processes.

sendbuf	send buffer array
sendoffset	initial offset in send buffer
recvbuf	receive buffer array
recvoffset	initial offset in receive buffer
recvcounts	numbers of result elements distributed to each process
datatype	data type of each item in send buffer
op	reduce operation

Java binding of the MPI operation MPI_REDUCE_SCATTER.

Scan

public void Scan(java.lang.Object sendbuf,
                 int sendoffset,
                 java.lang.Object recvbuf,
                 int recvoffset,
                 int count,
                 Datatype datatype,
                 Op op)

Perform a prefix reduction on data distributed across the group.

sendbuf	send buffer array
sendoffset	initial offset in send buffer
recvbuf	receive buffer array
recvoffset	initial offset in receive buffer
count	number of items in input buffer
datatype	data type of each item in input buffer
op	reduce operation

Java binding of the MPI operation MPI_SCAN.

Create_cart

public Cartcomm Create_cart(int[] dims,
                            boolean[] periods,
                            boolean reorder)

Create a Cartesian topology communicator whose group is a subset of the group of this communicator.

dims	the number of processes in each dimension
periods	true if grid is periodic, false if not, in each dimension
reorder	true if ranking may be reordered, false if not
returns:	new Cartesian topology communicator

Java binding of the MPI operation MPI_CART_CREATE.

The number of dimensions of the Cartesian grid is taken to be the size of the dims argument. The array periods must be the same size.

Create_graph

public Graphcomm Create_graph(int[] index,
                              int[] edges,
                              boolean reorder)

Create a graph topology communicator whose group is a subset of the group of this communicator.

index	node degrees
edges	graph edges
reorder	true if ranking may be reordered, false if not
returns:	new graph topology communicator

Java binding of the MPI operation MPI_GRAPH_CREATE.

The number of nodes in the graph, nnodes, is taken to be size of the index argument. The size of array edges must be index [nnodes} - 1].

GetGraph

public long GetGraph(int[] index,
                     int[] edges,
                     boolean reorder)