CSCI-4320/6360: Parallel Programming & Computing Tues./Fri. 12-1:30 p.m. MPI File I/O

PPC 2013 - MPI Parallel File I/O 1

CSCI-4320/6360: Parallel Programming & Computing

Tues./Fri. 12-1:30 p.m.MPI File I/O

Prof. Chris CarothersComputer Science Department

MRC [email protected]

www.cs.rpi.edu/~chrisc/COURSES/PARALLEL/SPRING-2013Adapted from: people.cs.uchicago.edu/~asiegel/courses/cspp51085/.../mpi-io.ppt


Common Ways of Doing I/O in Parallel Programs

• Sequential I/O:– All processes send data to rank 0, and 0 writes it to the

file


Pros and Cons of Sequential I/O

• Pros:– parallel machine may support I/O from only one

process (e.g., no common file system)– Some I/O libraries (e.g. HDF-4, NetCDF,

PMPIO) not parallel– resulting single file is handy for ftp, mv– big blocks improve performance– short distance from original, serial code

• Cons:– lack of parallelism limits scalability, performance

(single node bottleneck)


Another Way• Each process writes to a separate file

• Pros: – parallelism, high performance

• Cons: – lots of small files to manage

– LOTS OF METADATA – stress parallel filesystem

– difficult to read back data from different number of processes


What is Parallel I/O?• Multiple processes of a parallel

program accessing data (reading or writing) from a common file

FILE

P0 P1 P2 P(n-1)


Why Parallel I/O?• Non-parallel I/O is simple but

– Poor performance (single process writes to one file) or

– Awkward and not interoperable with other tools (each process writes a separate file)

• Parallel I/O– Provides high performance– Can provide a single file that can be used

with other tools (such as visualization programs)


Why is MPI a Good Setting for Parallel I/O?

• Writing is like sending a message and reading is like receiving.

• Any parallel I/O system will need a mechanism to– define collective operations (MPI

communicators)– define noncontiguous data layout in memory

and file (MPI datatypes)– Test completion of nonblocking operations (MPI

request objects)

• i.e., lots of MPI-like machinery


MPI-IO Background• Marc Snir et al (IBM Watson) paper

exploring MPI as context for parallel I/O (1994)

• MPI-IO email discussion group led by J.-P. Prost (IBM) and Bill Nitzberg (NASA), 1994

• MPI-IO group joins MPI Forum in June 1996

• MPI-2 standard released in July 1997

• MPI-IO is Chapter 9 of MPI-2


Using MPI for Simple I/O

FILE

P0 P1 P2 P(n-1)

Each process needs to read a chunk of data from a common file


Using Individual File Pointers#include<stdio.h>#include<stdlib.h>#include "mpi.h"#define FILESIZE 1000

int main(int argc, char **argv){

int rank, nprocs; MPI_File fh; MPI_Status status; int bufsize, nints; int buf[FILESIZE]; MPI_Init(&argc, &argv); MPI_Comm_rank(MPI_COMM_WORLD, &rank); MPI_Comm_size(MPI_COMM_WORLD, &nprocs); bufsize = FILESIZE/nprocs; nints = bufsize/sizeof(int); MPI_File_open(MPI_COMM_WORLD, "datafile", MPI_MODE_RDONLY, MPI_INFO_NULL, &fh); MPI_File_seek(fh, rank * bufsize, MPI_SEEK_SET); MPI_File_read(fh, buf, nints, MPI_INT, &status); MPI_File_close(&fh);}


Using Explicit Offsets

#include<stdio.h>#include<stdlib.h>#include "mpi.h"#define FILESIZE 1000

int main(int argc, char **argv){

int rank, nprocs; MPI_File fh; MPI_Status status; int bufsize, nints; int buf[FILESIZE]; MPI_Init(&argc, &argv); MPI_Comm_rank(MPI_COMM_WORLD, &rank); MPI_Comm_size(MPI_COMM_WORLD, &nprocs); bufsize = FILESIZE/nprocs; nints = bufsize/sizeof(int); MPI_File_open(MPI_COMM_WORLD, "datafile", MPI_MODE_RDONLY, MPI_INFO_NULL, &fh); MPI_File_read_at(fh, rank*bufsize, buf, nints, MPI_INT, &status); MPI_File_close(&fh);}


Function DetailsMPI_File_open(MPI_Comm comm, char *file, int mode, MPI_Info info, MPI_File *fh) (note: mode = MPI_MODE_RDONLY, MPI_MODE_RDWR, MPI_MODE_WRONLY, MPI_MODE_CREATE, MPI_MODE_EXCL,

MPI_MODE_DELETE_ON_CLOSE, MPI_MODE_UNIQUE_OPEN, MPI_MODE_SEQUENTIAL, MPI_MODE_APPEND)

MPI_File_close(MPI_File *fh)

MPI_File_read(MPI_File fh, void *buf, int count, MPI_Datatype type, MPI_Status *status)

MPI_File_read_at(MPI_File fh, int offset, void *buf, int count, MPI_Datatype type, MPI_Status *status)

MPI_File_seek(MPI_File fh, MPI_Offset offset, in whence); (note: whence = MPI_SEEK_SET, MPI_SEEK_CUR, or MPI_SEEK_END)

MPI_File_write(MPI_File fh, void *buf, int count, MPI_Datatype datatype, MPI_Status *status)MPI_File_write_at( …same as read_at … );(Note: Many other functions to get/set properties (see Gropp et al))


Writing to a File• Use MPI_File_write or MPI_File_write_at

• Use MPI_MODE_WRONLY or MPI_MODE_RDWR as the flags to MPI_File_open

• If the file doesn’t exist previously, the flag MPI_MODE_CREATE must also be passed to MPI_File_open

• We can pass multiple flags by using bitwise-or ‘|’ in C, or addition ‘+” in Fortran


MPI Datatype Interlude• Datatypes in MPI

– Elementary: MPI_INT, MPI_DOUBLE, etc• everything we’ve used to this point

• Contiguous– Next easiest: sequences of elementary types

• Vector– Sequences separated by a constant “stride”


MPI Datatypes, cont

• Indexed: more general– does not assume a constant stride

• Struct– General mixed types (like C structs)


Creating simple datatypes• Let’s just look at the simplest types:

contiguous and vector datatypes.• Contiguous example

– Let’s create a new datatype which is two ints side by side. The calling sequence is

MPI_Type_contiguous(int count, MPI_Datatype oldtype, MPI_Datatype *newtype);

MPI_Datatype newtype;

MPI_Type_contiguous(2, MPI_INT, &newtype);

MPI_Type_commit(newtype); /* required */


Using File Views

• Processes write to shared file

• MPI_File_set_view assigns regions of the file to separate processes


File Views• Specified by a triplet (displacement, etype, and filetype)

passed to MPI_File_set_view

• displacement = number of bytes to be skipped from the start of the file

• etype = basic unit of data access (can be any basic or derived datatype)

• filetype = specifies which portion of the file is visible to the process

• This is a collective operation and so all processors/ranks must use the same data rep, etypes in the group determined when the file was open..


File Interoperability

• Users can optionally create files with a portable binary data representation

• “datarep” parameter to MPI_File_set_view• native - default, same as in memory, not

portable• internal - impl. defined representation

providing an impl. defined level of portability• external32 - a specific representation defined in

MPI, (basically 32-bit big-endian IEEE format), portable across machines and MPI implementations


File View ExampleMPI_File thefile;

for (i=0; i<BUFSIZE; i++) buf[i] = myrank * BUFSIZE + i;

MPI_File_open(MPI_COMM_WORLD, "testfile", MPI_MODE_CREATE | MPI_MODE_WRONLY, MPI_INFO_NULL, &thefile);

MPI_File_set_view(thefile, myrank * BUFSIZE, MPI_INT, MPI_INT, "native",

MPI_INFO_NULL);

MPI_File_write(thefile, buf, BUFSIZE, MPI_INT, MPI_STATUS_IGNORE);

MPI_File_close(&thefile);


Ways to Write to a Shared File

• MPI_File_seek

• MPI_File_read_at

• MPI_File_write_at

• MPI_File_read_shared

• MPI_File_write_shared

• Collective operations

combine seek and I/Ofor thread safety

use shared file pointergood when order doesn’t matter

like Unix seek


Collective I/O in MPI

• A critical optimization in parallel I/O

• Allows communication of “big picture” to file system

• Framework for 2-phase I/O, in which communication precedes I/O (can use MPI machinery)

• Basic idea: build large blocks, so that reads/writes in I/O system will be large

Small individualrequests

Large collectiveaccess


Collective I/O• MPI_File_read_all, MPI_File_read_at_all, etc

• _all indicates that all processes in the group specified by the communicator passed to MPI_File_open will call this function

• Each process specifies only its own access information -- the argument list is the same as for the non-collective functions


Collective I/O

• By calling the collective I/O functions, the user allows an implementation to optimize the request based on the combined request of all processes

• The implementation can merge the requests of different processes and service the merged request efficiently

• Particularly effective when the accesses of different processes are noncontiguous and interleaved


Collective non-contiguousMPI-IO examples

#define “mpi.h”#define FILESIZE 1048576#define INTS_PER_BLK 16

int main(int argc, char **argv){ int *buf, rank, nprocs, nints, bufsize; MPI_File fh; MPI_Datatype filetype; MPI_Init(&argc, &argv); MPI_Comm_rank(MPI_COMM_WORLD, &rank); MPI_Comm_size(MPI_COMM_WORLD, &nprocs); bufsize = FILESIZE/nprocs; buf = (int *) malloc(bufsize); nints = bufsize/sizeof(int);

MPI_File_open(MPI_COMM_WORLD, “filename”, MPI_MODE_RD_ONLY, MPI_INFO_NULL, &fh); MPI_Type_vector(nints/INTS_PER_BLK, INTS_PER_BLK, INTS_PER_BLK*nprocs, MPI_INT, &filetype); MPI_Type_commit(&filetype); MPI_File_set_view(fh, INTS_PER_BLK*sizeof(int)*rank, MPI_INT, filetype, “native”, MPI_INFO_NULL); MPI_File_read_all(fh, buf, nints, MPI_INT, MPI_STATUS_IGNORE);

MPI_Type_free(&filetype); free(buf) MPI_Finalize(); return(0);}


More on MPI_Read_all• Note that the _all version has the same

argument list

• Difference is that all processes involved in MPI_Open must call this the read

• Contrast with the non-all version where any subset may or may not call it

• Allows for many optimizations


Split Collective I/O

MPI_File_write_all_begin(fh, buf, count, datatype);// available on Blue Gene/L, but may not improve // performancefor (i=0; i<1000; i++) { /* perform computation */}

MPI_File_write_all_end(fh, buf, &status);

• A restricted form of nonblocking collective I/O

• Only one active nonblocking collective operation allowed at a time on a file handle

• Therefore, no request object necessary


Passing Hints to the Implementation

MPI_Info info;

MPI_Info_create(&info);

/* no. of I/O devices to be used for file striping */MPI_Info_set(info, "striping_factor", "4");

/* the striping unit in bytes */MPI_Info_set(info, "striping_unit", "65536");

MPI_File_open(MPI_COMM_WORLD, "/pfs/datafile", MPI_MODE_CREATE | MPI_MODE_RDWR, info, &fh);

MPI_Info_free(&info);


Examples of Hints (used in ROMIO)

• striping_unit

• striping_factor

• cb_buffer_size

• cb_nodes

• ind_rd_buffer_size

• ind_wr_buffer_size

• start_iodevice

• pfs_svr_buf

• direct_read

• direct_write

MPI-2 predefined hints

New Algorithm Parameters

Platform-specific hints


I/O Consistency Semantics

• The consistency semantics specify the results when multiple processes access a common file and one or more processes write to the file

• MPI guarantees stronger consistency semantics if the communicator used to open the file accurately specifies all the processes that are accessing the file, and weaker semantics if not

• The user can take steps to ensure consistency when MPI does not automatically do so


Example 1• File opened with MPI_COMM_WORLD. Each process writes to a

separate region of the file and reads back only what it wrote.

MPI_File_open(MPI_COMM_WORLD,…)MPI_File_write_at(off=0,cnt=100)MPI_File_read_at(off=0,cnt=100)

MPI_File_open(MPI_COMM_WORLD,…)MPI_File_write_at(off=100,cnt=100)MPI_File_read_at(off=100,cnt=100)

Process 0 Process 1

• MPI guarantees that the data will be read correctly


Example 2• Same as example 1, except that each process

wants to read what the other process wrote (overlapping accesses)

• In this case, MPI does not guarantee that the data will automatically be read correctly

• In the above program, the read on each process is not guaranteed to get the data written by the other process!

/* incorrect program */MPI_File_open(MPI_COMM_WORLD,…)MPI_File_write_at(off=0,cnt=100)MPI_BarrierMPI_File_read_at(off=100,cnt=100)

/* incorrect program */MPI_File_open(MPI_COMM_WORLD,…)MPI_File_write_at(off=100,cnt=100)MPI_BarrierMPI_File_read_at(off=0,cnt=100)

Process 0 Process 1


Example 2 contd.• The user must take extra steps to ensure

correctness• There are three choices:

– set atomicity to true– close the file and reopen it– ensure that no write sequence on any

process is concurrent with any sequence (read or write) on another process/MPI rank• Can hurt performance….


Example 2, Option 1Set atomicity to true

MPI_File_open(MPI_COMM_WORLD,…)MPI_File_set_atomicity(fh1,1)MPI_File_write_at(off=0,cnt=100)MPI_BarrierMPI_File_read_at(off=100,cnt=100)

MPI_File_open(MPI_COMM_WORLD,…)MPI_File_set_atomicity(fh2,1)MPI_File_write_at(off=100,cnt=100)MPI_BarrierMPI_File_read_at(off=0,cnt=100)

Process 0 Process 1


Example 2, Option 2Close and reopen file

MPI_File_open(MPI_COMM_WORLD,…)MPI_File_write_at(off=0,cnt=100)MPI_File_closeMPI_BarrierMPI_File_open(MPI_COMM_WORLD,…)MPI_File_read_at(off=100,cnt=100)

MPI_File_open(MPI_COMM_WORLD,…)MPI_File_write_at(off=100,cnt=100)MPI_File_closeMPI_BarrierMPI_File_open(MPI_COMM_WORLD,…)MPI_File_read_at(off=0,cnt=100)

Process 0 Process 1


Example 2, Option 3• Ensure that no write sequence on any

process is concurrent with any sequence (read or write) on another process

• a sequence is a set of operations between any pair of open, close, or file_sync functions

• a write sequence is a sequence in which any of the functions is a write operation


Example 2, Option 3

MPI_File_open(MPI_COMM_WORLD,…)MPI_File_write_at(off=0,cnt=100)MPI_File_sync

MPI_Barrier

MPI_File_sync /*collective*/


MPI_Barrier

MPI_File_syncMPI_File_read_at(off=100,cnt=100)MPI_File_close

MPI_File_open(MPI_COMM_WORLD,…)


MPI_Barrier

MPI_File_syncMPI_File_write_at(off=100,cnt=100)MPI_File_sync

MPI_Barrier

MPI_File_sync /*collective*/MPI_File_read_at(off=0,cnt=100)MPI_File_close

Process 0 Process 1


General Guidelines for Achieving High I/O Performance

• Buy sufficient I/O hardware for the machine

• Use fast file systems, not NFS-mounted home directories

• Do not perform I/O from one process only

• Make large requests wherever possible

• For noncontiguous requests, use derived datatypes and a single collective I/O call


Optimizations

• Given complete access information, an implementation can perform optimizations such as:– Data Sieving: Read large chunks and

extract what is really needed– Collective I/O: Merge requests of

different processes into larger requests– Improved prefetching and caching


Summary• MPI-IO has many features that can help users

achieve high performance

• The most important of these features are the ability to specify noncontiguous accesses, the collective I/O functions, and the ability to pass hints to the implementation

• Users must use the above features!

• In particular, when accesses are noncontiguous, users must create derived datatypes, define file views, and use the collective I/O functions

CSCI-4320/6360: Parallel Programming & Computing Tues./Fri. 12-1:30 p.m. MPI File I/O

Documents

Transcript of CSCI-4320/6360: Parallel Programming & Computing Tues./Fri. 12-1:30 p.m. MPI File I/O