GPGPU programming with CUDA

Prepared by,
SAVITH.S
14CA65
MCA ,II SEM
NIT –K ,Surathkal
19/17/2015 Final Review Content 1

INTRODUCTION
 GPU- stands for Graphics processing unit (visual
processing unit (VPU)) ,simply our graphics card.
 Electronic circuit that is used to accelerate the creation
of images on frame buffer and enhances the output
quality.
 Generally interact with motherboard through
PCIexpress (PCIe) port or AGP ports.
 Very efficient tool for manipulating computer
graphics.
 Today, parallel GPUs have begun making
computational inroads against the CPU.
 It have wide range of options due to its parallel
behavior ,it leads to the development of GPGPU.

ABSTRACT
GPGPU stands for general purpose graphical processing
unit , that is designed to use the GPU for algorithms
that are traditionally run on the CPU. It makes the
algorithms very faster in execution and it saves the
processing time. So there is unlimited applications are
possible by the usage of this concept. Any GPU
providing a functionally complete set of operations
performed on arbitrary bits can compute any
computable value. Additionally, the use of multiple
graphics cards in one computer, or large numbers of
graphics chips, further parallelizes the already parallel
nature of graphics processing. So by customizing a
GPU we can implement a GPGPU which is hundred
times faster than a traditional cpu.

GPGPU
 Utilization of GPU for various computations which is
traditionaly handled by the CPU.
 Can perform any set of operations very accurately and
can compute any computable value.
 use of multiple graphics cards in one computer, or
large numbers of graphics chips, further parallelizes
the already parallel nature of graphics processing.
 This concept turns the massive computational power
of a modern graphics accelerator's shader pipeline into
general-purpose computing power .

STREAM PROCESSING
 STREAM is a set of records that need similar type of
computations .
 In traditional GPU we can read multiple independent
rows simultaneously perform operations and write
multiple outputs ,But never have a piece of m/y ie,
Both readable and writable.
 Utilization of massive computational power of a gpu
into the general purpose cpu related operations.
 Generalization of GPU.

CUDA
 CUDA stands for Compute Unified Device
Architecture devaloped by NVIDIA cooperation.
 It is used to develop software for graphics processors
and is used to develop a variety of general purpose
applications for GPUs that are highly parallel in nature
and run on hundreds of GPU’s processor cores.
 CUDA is supported only on NVIDIA’s GPUs based on
Tesla architecture. The graphics cards that support
CUDA are GeForce 8-series, Quadro, and Tesla.
 CUDA has some specific functions, called kernels. It is
executed N number of times in parallel on GPU by
using N number of threads.

EXISTING SYSTEM
 The existing system may be a traditional GPU or a
CPU.
 A CPU is used in normal case and which has less
performance power and throughput when compare to
the GPU also having less parallel nature.
 A GPU is the basic form of GPGPU but it is used only
for graphic accelerating options such as game consoles
high definition images , computer aided designs etc .

PROPOSED SYSTEM
 The GPGPU solves problems of a traditional CPU by
its highly parallel nature .
 In principle, any Boolean function can be built-up
from a functionally complete set of logic operators.
 GPGPU applications to have high arithmetic intensity
else the memory access latency will limit
computational speed up .
 Ideal GPGPU applications have large data sets, high
parallelism, and minimal dependency between data
elements.

EXPECTED FUNCTIONALITIES
 Having a high performance nature and about hundred
times faster than traditional CPU.
 Highly parallel behavior .
 Contains multiple cores , each has individual executing
nature.
 Implementing multiple GPU’s in a single system improves
its functionalities again and again .
 Can be customized to any type by using different platforms
like CUDA ,openCL etc.
 Major upcoming applications in high performance
computing areas.

WORKING

SYSTEM IMPLEMENTATION
 The implementation of system is completely carried
out by the hardware assembling of GPU in a
normal/special purpose computer system and software
installation .
 There is no separate software s needed for customizing
a GPU . It can be done by the CUDA C or CUDA C++
with NVIDIA’s compiler .
 The programming is in the usual way but we have to
implement some additional header files like #include
mpich.h (explained in the later section) in the
program .

HARDWARE REQUIREMENTS
 GPU shader cores, which run GPU kernels, are both
parallel and deeply multithreaded to provide
significant computational power, currently on the
order of a tera flopper GPU.
 Graphics memory, which is directly accessible by GPU
kernels, has a high clock rate and wide bus width to
provide substantial bandwidth, currently about a
hundred gigabytes per second.
 GPU interconnect, providing main board access to the
GPU. This is typically PCI Express, and so delivers afew
gigabytes per second of bandwidth.

 Main board RAM, which is directly accessible by CPU
programs and the network.
 CPU cores, which are deeply pipelined and
superscalarto provide good performance on sequential
programs.
 Network hardware, which moves bytes between
nodes. We use the trivial latency plus bandwidth
performance.

THE GRID AND BLOCK STRUCTURE
 The Grid consists of one-dimensional, two-
dimensional or three-dimensional thread blocks.
 Each thread block is further divided into one-
dimensional or two-dimensional threads.
 A thread block is a set of threads running on one
processor.
 All this threads creation, their execution, and
termination are automatic and handled by the GPU,
and is invisible to the programmer.
 The user only needs to specify the number of threads
in a thread block and the number of thread blocks in a
grid.

SINGLE PROGRAM MULTIPLE DATA
(SPMD) & MPICH
 GPU is suited for single program multiple data type
parallel calculations and work well with message
passing interface approach to programming .
 In SPMD concept , There is only a single program for
controlling various activities done in the GPU .
 There is no direct connection between our network
device and GPU memory.
 Thus to send GPU data across the network, we must
copy the send-side GPU data to CPU memory .
 So we are using a standard CPU interface such as MPI,
and finally copy the received data from CPU memory
into GPU memory.

 MPICH is a freely available, portable implementation
of MPI, a standard for message-passing for
distributed-memory applications used in parallel
computing’ .
 The CH part of the name was derived from
"Chameleon", which was a portable parallel
programming library developed by William Gropp,
one of the founders of MPICH.
 After installing MPICH they we have to create a user
using user add or the GUI. Also set password
MPD_SECRETWORD=password
 Here password refers to the password given for their
user id.

 Next change the read/write/execute profile of
.mpd.conf using chmod 600.mpd.conf . Next create a
file named mpd.hosts containing the following :
Master
Node1
Node2
..
Node m-1
Where m in node m-1 refers to the total number of
nodes .
 Next to boot MPICH type :
Mpdboot –n m –r ssh –f mpd.hosts

Given below is a sample program using mpi in cuda .
#include<stdio.h>
#include<mpi.h>
Main(int argc,char **,argv){
Int I,root=0,rank,size,tag=32,mysum,total;
Mpi_init(&argc,&argv);
/*gets rank or identity of each processor*/
Mpi_Comm_rank(MPI_COMM_WORLD,&rank);
/*gets the total number of available processors*/
Mpi_Comm_size(MPI_COMM_WORLD,&size);
Mysum=0;
Total=0;
For(i=rank+1;i<100;i=i+size)
mysum=mysum+1;
/*Adds all the partial sums called mysum and stores into total at root using the MPI_SUM
call */
MPI_Reduce(&mysum,&total,tag,MPI_INT,MPI_SUM,root,
MPI_INT,MPI_SUM,root,MPI_COMM_WORLD)
If(rank==0)
Printf(“The total is %dn”,total);
}
In this program mp processors add upto 100. The first processor adds 1,mp+1,2*mp+1 ; the
second processor adds 2,mp+2,2*mp+2 ; etc…..

APPLICATIONS
Research: Higher Education and Supercomputing.
Computational Chemistry and Biology.
Bioinformatics
Molecular Dynamics
High Performance Computing (HPC) clusters.
Grid computing.
Auto signal processing.
Scientific computing.

CONCLUSION AND FUTURE WORK
 It is clear that by using GPGPU we can execute many
rows of data in parallel thus it provides a high
performance .
 The NVIDIA’s cuda is well suited for building the
GPGPU platform.
 We have presented and benchmarked cudaMPI and
glMPI, message passing libraries for distributed-
memory GPU clusters.
 Many functions and variables of MPI is under
development .
 A common platform for various GPU’s is needed and is
under development .

Interactive Section

GPGPU programming with CUDA

More Related Content

What's hot

Viewers also liked

Similar to GPGPU programming with CUDA

Recently uploaded

GPGPU programming with CUDA