Transcript ppt - MPJ Express

Parallel Programming with Java
Aamir Shafi
National University of Sciences and
Technology (NUST)
http://hpc.seecs.edu.pk/~aamir
http://mpj-express.org
1
Two Important Concepts
• Two fundamental concepts of parallel
programming are:
• Domain decomposition
• Functional decomposition
2
Domain Decomposition
Image taken from https://computing.llnl.gov/tutorials/parallel_comp/
3
Functional Decomposition
Image taken from https://computing.llnl.gov/tutorials/parallel_comp/
4
Message Passing Interface (MPI)
• MPI is a standard (an interface or an API):
•
•
•
•
It defines a set of methods that are used by application developers to write their applications
MPI library implement these methods
MPI itself is not a library—it is a specification document that is followed!
MPI-1.2 is the most popular specification version
• Reasons for popularity:
•
•
•
•
Software and hardware vendors were involved
Significant contribution from academia
MPICH served as an early reference implementation
MPI compilers are simply wrappers to widely used C and Fortran compilers
• History:
• The first draft specification was produced in 1993
• MPI-2.0, introduced in 1999, adds many new features to MPI
• Bindings available to C, C++, and Fortran
• MPI is a success story:
• It is the mostly adopted programming paradigm of IBM Blue Gene systems
• At least two production-quality MPI libraries:
• MPICH2 (http://www-unix.mcs.anl.gov/mpi/mpich2/)
• OpenMPI (http://open-mpi.org)
• There’s even a Java library:
• MPJ Express (http://mpj-express.org)
5
Message Passing Model
• Message passing model allows processors to
communicate by passing messages:
• Processors do not share memory
• Data transfer between processors required cooperative
operations to be performed by each processor:
• One processor sends the message while other receives the
message
6
Distributed Memory Cluster
barq.niit.edu.pk
(cluster head node)
chenab1
chenab2
chenab3
chenab7
chenab6
chenab4
chenab5
7
Steps involved in executing the
“Hello World!” program
1.
2.
3.
4.
5.
6.
7.
Let’s logon to the cluster head node
Write the Hello World program
Compile the program
Write the machines files
Start MPJ Express daemons
Execute the parallel program
Stop MPJ Express daemons
8
Step1: Logon to the head node
9
Step 2: Write the Hello World
Program
10
Step 3: Compile the code
11
Step 4: Write the machines file
12
Step 5: Start MPJ Express
daemons
13
Step 6: Execute the parallel
program
aamir@barq:~/projects/mpj-user> mpjrun.sh -np 6 -headnodeip 10.3.20.120 dport 11050 HelloWorld
..
Hi from process <3> of total <6>
Hi from process <1> of total <6>
Hi from process <2> of total <6>
Hi from process <4> of total <6>
Hi from process <5> of total <6>
Hi from process <0> of total <6>
…
14
Step 7: Stop the MPJ Express
daemons
15
COMM WORLD Communicator
import java.util.*;
import mpi.*;
..
// Initialize MPI
MPI.Init(args); // start up MPI
// Get total number of processes and rank
size = MPI.COMM_WORLD.Size();
rank = MPI.COMM_WORLD.Rank();
..
16
What is size?
import java.util.*;
import mpi.*;
..
// Get total number of processes
size = MPI.COMM_WORLD.Size();
..
• Total number of processes in a communicator:
• The size of MPI.COMM_WORLD is 6
17
What is rank?
import java.util.*;
import mpi.*;
..
// Get total number of processes
rank = MPI.COMM_WORLD.Rank();
..
• The “unique” identify (id) of a process in a communicator:
• Each of the six processes in MPI.COMM_WORLD has a distinct rank
or id
18
Single Program Multiple Data
(SPMD) Model
import java.util.*;
import mpi.*;
public class HelloWorld {
MPI.Init(args); // start up MPI
size = MPI.COMM_WORLD.Size();
rank = MPI.COMM_WORLD.Rank();
if (rank == 0) {
System.out.println(“I am Process 0”);
}
else if (rank == 1) {
System.out.println(“I am Process 1”);
}
MPI.Finalize();
}
19
Single Program Multiple Data
(SPMD) Model
import java.util.*;
import mpi.*;
public class HelloWorld {
MPI.Init(args); // start up MPI
size = MPI.COMM_WORLD.Size();
rank = MPI.COMM_WORLD.Rank();
if (rank%2 == 0) {
System.out.println(“I am an even process”);
}
else if (rank%2 == 1) {
System.out.println(“I am an odd process”);
}
MPI.Finalize();
}
20
Point to Point Communication
• The most fundamental facility provided by MPI
• Basically “exchange messages between two
processes”:
• One process (source) sends message
• The other process (destination) receives message
21
Point to Point Communication
• It is possible to send message for each basic datatype:
• Floats (MPI.FLOAT), Integers (MPI.INT), Doubles
(MPI.DOUBLE) …
• Java Objects (MPI.OBJECT)
• Each message contains a “tag”—an identifier
Tag1
Tag2
22
Point to Point Communication
Process 1
Process 2
Process 0
message
Integers
Process 4
Tag COMM_WORLD
Process 3
Process 7
Process 6
Process 4
Process 5
23
Blocking Send() and Recv()
Methods
public void Send(Object buf, int offset, int count,
Datatype datatype, int dest, int tag) throws MPIException
public Status Recv(Object buf, int offset, int count,
Datatype datatype, int src, int tag) throws MPIException
24
Blocking and Non-blocking
Point-to-Point Comm
• There are blocking and non-blocking version of send
and receive methods
• Blocking versions:
• A process calls Send() or Recv(), these methods return when
the message has been physically sent or received
• Non-blocking versions:
• A process calls Isend() or Irecv(), these methods return
immediately
• The user can check the status of message by calling Test() or
Wait()
• Non-blocking versions provide overlapping of
computation and communication:
• Asynchronous communication
25
Sender
“Blocking”
time
Send()
Recv()
CPU waits
Sender
Isend()
time
Receiver
CPU does
computation
Wait()
CPU waits
CPU waits
“Non Blocking”
Receiver
Irecv()
CPU does
computation
Wait()
CPU waits
26
Non-blocking Point-to-Point
Comm
public Request Isend(Object buf, int offset, int count,
Datatype datatype, int dest, int tag) throws MPIException
public Request Irecv(Object buf, int offset, int count,
Datatype datatype, int src, int tag) throws MPIException
public Status Wait() throws MPIException
public Status Test() throws MPIException
27
Performance Evaluation of Point to
Point Communication
• Normally ping pong benchmarks are used to
calculate:
• Latency: How long it takes to send N bytes from
sender to receiver?
• Throughput: How much bandwidth is achieved?
• Latency is a useful measure for studying the
performance of “small” messages
• Throughput is a useful measure for studying the
performance of “large” messages
28
Latency on Myrinet
29
Throughput on Myrinet
30
Collective communications
• Provided as a convenience for application
developers:
• Save significant development time
• Efficient algorithms may be used
• Stable (tested)
• Built on top of point-to-point communications
• These operations include:
• Broadcast, Barrier, Reduce, Allreduce, Alltoall,
Scatter, Scan, Allscatter
• Versions that allows displacements between the data
31
Broadcast, scatter, gather,
allgather, alltoall
32
Image from MPI standard doc
Broadcast, scatter, gather,
allgather, alltoall
public void Bcast(Object buf, int offset, int count,
Datatype type, int root) throws MPIException
public void Scatter(Object sendbuf, int sendoffset, int sendcount, Datatype sendtype,
Object recvbuf, int recvoffset, int recvcount, Datatype recvtype,
int root) throws MPIException
public void Gather(Object sendbuf, int sendoffset, int sendcount, Datatype sendtype,
Object recvbuf, int recvoffset, int recvcount, Datatype recvtype,
int root) throws MPIException
public void Allgather(Object sendbuf, int sendoffset int sendcount, Datatype sendtype,
Object recvbuf, int recvoffset, int recvcount, Datatype recvtype)
throws MPIException
public void Alltoall(Object sendbuf, int sendoffset, int sendcount, Datatype sendtype,
Object recvbuf, int recvoffset, int recvcount, Datatype recvtype)
throws MPIException
33
Reduce collective operations
Processes
Processes
Data
1
15
2
3
reduce
4
5
1
15
2
15
3
allreduce
15
4
15
5
15












MPI.PROD
MPI.SUM
MPI.MIN
MPI.MAX
MPI.LAND
MPI.BAND
MPI.LOR
MPI.BOR
MPI.LXOR
MPI.BXOR
MPI.MINLOC
MPI.MAXLOC
34
Reduce collective operations
public void Reduce(Object sendbuf, int sendoffset,
Object recvbuf, int recvoffset, int count,
Datatype datatype, Op op, int root)
throws MPIException
public void Allreduce(Object sendbuf, int sendoffset,
Object recvbuf, int recvoffset, int count,
Datatype datatype, Op op)
throws MPIException
35
Collective Communication
Performance
36
Summary
• MPJ Express is a Java messaging system that
can be used to write parallel applications:
• MPJ/Ibis and mpiJava are other similar software
• MPJ Express provides point-to-point
communication methods like Send() and Recv():
• Blocking and non-blocking versions
• Collective communications is also supported
• Feel free to contact me if you have any queries
37