1 / 26

CSS434 Group Communication and MPI Textbook Ch4.4-5 and 15.4

CSS434 Group Communication and MPI Textbook Ch4.4-5 and 15.4. Professor: Munehiro Fukuda. Outline. Reliability of group communication IP multicast Atomic multicast Ordering of group communication Absolute ordering Total ordering Causal ordering FIFO ordering MPI Java

lewis
Télécharger la présentation

CSS434 Group Communication and MPI Textbook Ch4.4-5 and 15.4

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. CSS434 Group Communication and MPI Textbook Ch4.4-5 and 15.4 Professor: Munehiro Fukuda CSS434 MPI

  2. Outline • Reliability of group communication • IP multicast • Atomic multicast • Ordering of group communication • Absolute ordering • Total ordering • Causal ordering • FIFO ordering • MPI Java • Programming, compilation and invocation • Major group communication functions: Bcast( ), Reduce( ), Allreduce( ) CSS434 MPI

  3. Group Communication • Communication types • One-to-many: broadcast • Many-to-one: synchronization, collective communication • Many-to-many: gather and scatter • Applications • Fault tolerance based on replicated services (type: one-to-many) • Finding the discovery servers in spontaneous networking (type: a one-to-many request, a many-to-one response) • Better performance through replicated data (type: one-to-many, many-to-one) • Propagation of event notification (type: one-to-many) CSS434 MPI

  4. IP (Unreliable) Multicast import java.net.*; import java.io.*; public class MulticastPeer{ public static void main(String args[]){ // args give message contents & destination multicast group (e.g. "228.5.6.7") MulticastSocket s =null; try { InetAddress group = InetAddress.getByName(args[1]); s = new MulticastSocket(6789); s.joinGroup(group); byte [] m = args[0].getBytes(); DatagramPacket messageOut = new DatagramPacket(m, m.length, group, 6789); s.send(messageOut); // get messages from others in group byte[] buffer = new byte[1000]; for(int i=0; i< 3; i++) { DatagramPacket messageIn = new DatagramPacket(buffer, buffer.length); s.receive(messageIn); System.out.println("Received:" + new String(messageIn.getData())); } s.leaveGroup(group); }catch (SocketException e){System.out.println("Socket: " + e.getMessage()); }catch (IOException e){System.out.println("IO: " + e.getMessage());} }finally {if(s != null) s.close();} } } Using a special network address: IP Class D and UDP CSS434 MPI

  5. Reliability and Ordering • Fault tolerance based on replicated services • Send-to-all and all-reliable semantics • Finding the discovery servers in spontaneous networking • 1-reliable semantics • Better performance through replicated data • Semantics depend on applications. (send-to-all and all- or m-out-of-n reliable semantics) • Propagation of event notifications • In general, send-to-all and all-reliable semantics CSS434 MPI

  6. Atomic Multicast • Send-to-all semantics and all-reliable • Simple emulation: • A repetition of one-to-one communication with acknowledgment • What if a receiver fails • Time-out retransmission • What if a sender fails before all receivers receive a message • All receivers forward the message to the same group and thereafter deliver it to themselves. • A receiver discard the 2nd or the following messages. CSS434 MPI

  7. Message Ordering • R1 and R2 receive m1 and m2 in a different order! • Some message ordering required • Absolute ordering • Consistent/total ordering • Causal ordering • FIFO ordering S2 S1 R1 R2 m2 m1 m1 m2 CSS434 MPI

  8. Absolute Ordering • Rule: • Mi must be delivered before mj if Ti < Tj • Implementation: • A clock synchronized among machines • A sliding time window used to commit message delivery whose timestamp is in this window. • Example: • Distributed simulation • Drawback • Too strict constraint • No absolute synchronized clock • No guarantee to catch all tardy messages Ti < Tj Ti mi Tj mi mj mj CSS434 MPI

  9. Consistent/Total Ordering • Rule: • Messages received in the same order (regardless of their timestamp). • Implementation: • A message sent to a sequencer, assigned a sequence number, and finally multicast to receivers • A message retrieved in incremental order at a receiver • Example: • Replicated database updates • Drawback: • A centralized algorithm Ti < Tj Ti Tj mj mj mi mi CSS434 MPI

  10. Total Ordering Using A Sequencer Message sent to all group members and a sequencer Receive an incoming message in a temporary queue Receive a sequence number message from the sequencer, Reorder the incoming message with this sequence #, and Deliver it if my local counter reaches this number. Receive a message, associate it with a sequence number. multicast it to all group members, and increments the sequence number CSS434 MPI

  11. The ISIS System for Total Ordering Ap4 Proposed Seq Pp4= max(Ap4, Pp4) + 1 P Pp4 2 1 Message Receiver Receiver 3 Ap4 2 P 2 4 2 Proposed Seq Pp4 1 Ap4= max(Ap4, a) 3 Agreed Seq a = max(Pp1, Pp2, P p3, P p4) 1 2 P Sender 1 3 Receiver Ap4 P 3 CSS434 MPI Pp4

  12. Causal Ordering • Rule: • Happened-before relation • If eki, eli ∈h and k < l, then eki → eli, • If ei = send(m) and ej = receive(m), then ei → ej, • If e → e’ and e’ → e”, then e → e” • Implementation: • Use of a vector message • Example: • Distributed file system • Drawback: • Vector as an overhead • Broadcast assumed R1 S1 R2 S2 R3 m4 m1 m1 m4 m2 m2 m3 From R2’s view point m1 →m2 CSS434 MPI

  13. Causal Ordering Using A Vector Stamps Each process maintains a vector. Increment my vector element, and Send a messge with a vector Make sure J’s element of J’s vector reaches J’s element of my vector. (All J’s previous message has been delivered.) Make sure all elements of J’s vector <= all elements of my vector. (I’ve delivered all messages that J has delivered.) Increment the sender’s vector element CSS434 MPI

  14. Vector Stamps • S[i] = R[i] + 1 where i is the source id • S[j] ≤ R[j] where i≠j Site B Site C Site D Site A 1, 1, 1, 0 2, 1, 0, 0 2, 1, 1, 0 2, 1, 1, 0 3,1,1,0 delayed delayed delivered CSS434 MPI

  15. FIFO Ordering • Rule: • Messages received in the same order as they were sent. • Implementation: • Messages assigned a sequence number • Example: • TCP • This is the weakest ordering. S R m1 Router 1 m2 m1 m3 m2 m4 m3 Router 2 m4 CSS434 MPI

  16. Why High-Level Message Passing Tools? • Data formatting • Data formatted into appropriate types at user level • Non-blocking communication • Polling and interrupt handled at system call level • Process addressing • Inflexible hardwired addressing with machine id + local id • Group communication • Group server implemented at user level • Broadcasting simulated by a repetition of one-to-one communication CSS434 MPI

  17. PVM and MPI • PVM: Parallel Virtual Machine • Developed in 80’s • The pioneer library to provide high-level message passing functions • The PVM daemon process taking care of message transfer for user processes in background • MPI: Message Passing Interface • Defined in 90’s • The specification of high-level message passing functions • Several implementations available: mpich, mpi-lam • Library functions directly linked to user programs (no background daemons) • The detailed difference is shown by: • PVMvsMPI.pdf CSS434 MPI

  18. Getting Started with MPI Java • Website: http://www.hpjava.org/courses/arl/lectures/mpi.ppt http://www.hpjava.org/reports/mpiJava-spec/mpiJava-spec.pdf • Creating a machines file: [mfukuda@UW1-320-00 mfukuda]$ vi machines uw1-320-00 uw1-320-01 uw1-320-02 uw1-320-03 • Compile a source program: [mfukuda@UW1-320-00 mfukuda]$ javac MyProg.java • Run the executable file: [mfukuda@UW1-320-00 mfukuda]$ prunjava 4 MyProg args CSS434 MPI

  19. Program Using MPI import mpi.*; class MyProg { public static void main( String[] args ) { MPI.Init( args ); // Start MPI computation int rank = MPI.COMM_WORLD.Rank( ); // Process ID (from 0 to #processes – 1) int size = MPI.COMM_WORLD.Size( ); // # participating processes System.out.println( "Hello World! I am " + rank + " of " + size ); MPI.Finalize(); // Finish MPI computation } } CSS434 MPI

  20. MPI_Send and MPI_Recv void MPI.COMM_WORLD.Send( Object[] message /* in */, int offset /* in */, int count /* in */, MPI.Datatype datatype /* in */, int dest /* in */, int tag /* in */) Status MPI.COMM_WORLD.Recv( Object[] message /* in */, int offset /* in */, int count /* in */, MPI.Datatype datatype /* in */, int source /* in */, int tag /* in */) int Status.Get_count( MPI.Datatype, datatype ) /* #objects received */ MPI.Datatype = BYTE, CHAR, SHORT, INT, LONG, FLOAT, DOUBLE, OBJECT CSS434 MPI

  21. MPI.Send and MPI.Recv import mpi.*; class myProg { public static void main( String[] args ) { int tag0 = 0; MPI.Init( args ); // Start MPI computation if ( MPI.COMM_WORLD.rank() == 0 ) { // rank 0…sender int loop[1]; loop[0] = 3; MPI.COMM_WORLD.Send( "Hello World!", 12, MPI.CHAR, 1, tag0 ); MPI.COMM_WORLD.Send( loop, 1, MPI.INT, 1, tag0 ); } else { // rank 1…receiver int loop[1]; char msg[12]; MPI::COMM_WORLD.Recv( msg, 12, MPI.CHAR, 0, tag0 ); MPI::COMM_WORLD.Recv( loop, 1, MPI.INT, 0, tag0 ); for ( int i = 0; i < loop[0]; i++ ) System.out.println( msg ); } MPI.Finalize( ); // Finish MPI computation } } CSS434 MPI

  22. Message Ordering in MPI • FIFO Ordering in each data type • Messages reordered with a tag in each data type Source Destination Source Destination tag = 1 tag = 3 tag = 2 CSS434 MPI

  23. MPI.Bcast void MPI.COMM_WORLD.Bcast( Object[] message /* in */, int offset /* in */, int count /* in */, MPI.Datatype datatype /* in */, int root /* in */) msg[0] Rank 3 Rank 2 Rank 4 Rank 1 Rank 0 MPI::COMM_WORLD.Bcast( msg, 0, 1, MPI.INT, 2); CSS434 MPI

  24. Rank3 8 Rank2 12 Rank4 4 Rank1 10 Rank0 15 49 MPI_Reduce void MPI.COMM_WORLD.Reduce( Object[] sendbuf /* in */, int sendoffset /* in */, Object[] recvbuf /* out */, int recvoffset /* in */, int count /* in */, MPI.Datatype datatype /* in */, MPI.Op operator /* in */, int root /* in */ ) MPI.Op = MPI.MAX (Maximum), MPI.MIN (Minimum), MPI.SUM (Sum), MPI.PROD (Product), MPI.LAND (Logical and), MPI.BAND (Bitwise and), MPI.LOR (Logical or), MPI.BOR (Bitwise or), MPI.LXOR (logical xor), MPI.BXOR(Bitwise xor), MPI.MAXLOC (MAX location) MPI.MINLOC (MIN loc.) MPI::COMM_WORLD.Reduce( msg, 0, result, 0, 1, MPI.INT, MPI.SUM, 2); CSS434 MPI

  25. 0 1 2 4 3 5 6 7 0 1 2 4 3 5 6 7 0 1 2 4 3 5 6 7 0 1 2 4 3 5 6 7 MPI_Allreduce void MPI.COMM_WORLD.Allreduce( Object[] sendbuf /* in */, int sendoffset /* in */, Object[] recvbuf /* out */, int recvoffset /* in */, int count /* in */, MPI.Datatype datatype /* in */, MPI.Op operator /* in */) CSS434 MPI

  26. Exercises (No turn-in) • Explain, in atomic multicast on page 6, why reversing the order of operations “all receivers forward the message to the same group and thereafter deliver it to themselves” makes the multicast no longer atomic. • Assume that four processes communicate with one another in causal ordering. Their current vectors are show below. If Process A sends a message, which processes can receive it immediately? • Show that, if the basic multicast that we use in the algorithm of P9 is also FIFO-ordered, then the resultant totally-ordered multicast is also causally ordered. • Consider pros and cons of PVM’s daemon-based and MPI’s library linking-based message passing. • Why can MPI maintain FIFO ordering? CSS434 MPI

More Related