C-DAC Certified HPC Professional Certification Program (CCHPCP):

With the increasing demand for skilled professionals in High Performance Computing (HPC) and the current dearth of qualified manpower, HPC Solutions Group, Centre for Development of Advanced Computing (C-DAC), Pune has announced the launch of C-DAC Certified HPC Professional Certification Program (CCHPCP). The primary objective of this program is to offer a standardized scale to map the HPC skillsets of both current and aspiring professionals in this domain, and enable them to pursue careers in HPC. This will also assist the Government, R&D and IT industry to recruit the right candidates from a pool of HPC industry ready and certified professionals.

Exam Information:

Exam Date: 1st September 2012

Registration starts by 25th June 2012

Last date of Registration 25th August 2012

C-DAC Certified HPC Professional Certification Program (CCHPCP) examination will be available online. This online certification exam initially will be conducted every three months and in due course, on-demand 24 hours a day, seven days a week through the existing Advanced Computing Training School (ACTS) Authorised Training Centres (ATC) network located across India acting as prometric centers. Registration can be done at the existing C-DAC ACTS ATC. For detailed list of Centres refer to acts.cdac.in Once the exam fees are paid it will be on non-refundable basis.

CCHPCP online exam will be a closed-book exam where books, course materials and notes will not be allowed. Candidates who take the online exams are subject to C-DAC norms and policies.

To be admitted into an examination center, each candidate must present a valid government issued photo identification with the candidates photograph (e.g., driver’s licence, passport, etc.). Each candidate will be required to sign in at the examination center. A candidate who does not present positive identification or who refuses or is unable to provide a matching signature will not be permitted to write the examination.

Registration / How to Apply

Register Online

Certification Exam Fees:

Rs.12,500/- (Rupees Twelve Thousand Five Hundred only) for Indian citizens.
US $ 500/- (Five Hundred Dollars only) for Foreign nationals.

Arriving at the Prometric Testing center:

• Reporting at the testing center at least 30 minutes prior to the scheduled appointment is compulsory. Prior to entering the testing room, candidates will be identified and processed for admission by the Test Administrator. The candidate must be prepared to show at least two forms of identification. One must be a government-issued photo identification. Both forms of identification must contain candidates signature.
• If the candidate reports more than 15 minutes late, the testing center personnel may choose not to seat the candidate, if doing so disrupts the other examinees in the testing room.
• The candidate is not permitted to bring papers, books, bags, computers, personal organizers, or calculators into the testing room. The testing center personnel will provide you with an erasable noteboard or paper for use while testing. All noteboards and paper will be collected by testing center personnel at the end of the exam.

Questions and Timing:

The examination will comprise of 150 objective multiple-choice type questions with no negative marking for wrong answers. The total exam time will be of 90 minutes (One and half hours). The required percentage for passing would be 70%.

Test Results:

Exam results will available at the testing center upon completion of the test on the same day. Certification from C-DAC shall be provided to the successful candidates.

Syllabus*:

Syllabus for the C-DAC Certified HPC Professional Certification Program (CCHPCP) is as under:

C, C++, FORTRAN and Computer Fundamentals
	C and C++
		Preprocessor Directives, Functions, pointers, arrays, structures
		Data Structures such as stacks, trees, graphs
		Algorithm implementation such as searching, sorting

	FORTRAN
		FORTRAN77
		FORTRAN90
		FORTRAN95
		High Performance FORTRAN (HPF)
		Subroutines, Functions, Modules, Interface
		File Handling
		Arrays, Pointers, Derived Data Types
		Linked Lists Program Units
		Debugging
		Use of Program Libraries

	Uni-Processor Optimization of Serial Codes
	Overview of Parallel Computing
		Why Parallel Computing?
			What is Parallel Computing?
			Parallel Programming Models
			Parallel Algorithmic Paradigms
			Classification Of Parallel Architectures
			HPC Clusters
			Scalability of parallel programs
			Performance Metrics
			Problems of Parallelism
		Issues and Challenges

	OS Concepts, Linux and Cluster Building and Administration
		OS Concepts
		OS Structures Process Management Memory Management Distributed Systems Special Purpose Systems
		Linux Environment
		Installation of Linux Directory Structure and Predefined Environment Variables Essential tools (grep, tar, yum, links, permissions etc.) Shell Scripting Linux Partitions Logical Volume Manager (LVM) Administration of users and groups Linux scheduling and services
		Cluster Building and Administration
		HPC Architecture HPC Cluster Components (Hardware and Software) Cluster Management Suite e.g. ROCKS, Oscar, xCat Workload and Resource Managers Virtualization Cluster Monitoring Tools, e.g. Ganglia, Nagios and Cacti Installing Compilers & License Manager Storage (RAID, NAS, SAN and NFS) Parallel File system(GFS, GPFS, Luster) Back-up Management

	Profiling and Debugging of Codes
		Identify opportunities for parallelism
		Performance Metrics of codes
		Debugging Tools
		Debugging of codes

	Message Passing Interface
		Basic MPI
			MPI Point - to Point
			MPI Collective Communication
				Data Synchronization
				Data Movement
				Collective Computation
	Advance MPI
		Derived Data Types
			Derived Types
			Special type Constructors
			Type Matching
			Packing/Unpacking Data
		Groups and Communicators
			Virtual Topologies
		MPI 2 Standard
			Parallel I/O
			Dynamic Process Spawning
			Interprocess Communication
		Case Studies (Algorithms and Parallelization Approaches)
			Jacobian Solver
			Monte Carlo Algorithm
			Matrix - Matrix Multiplication (Cannon’s Algorithm)
			FFT

	OpenMP
		Why OpenMP
		OpenMP Programming Model
		OpenMP constructs
		Case-Studies (Algorithms and Parallelization Approaches)
			Matrix -Matrix-multiplication
			Density Matrix Renormalization Group Algorithm

	Pthreads
		Pthreads Overview
			What is a thread?
			What are Pthreads?
			Why Pthreads?
			Designing Threaded Programs
		Pthreads APIs
		Thread Management
			Creating and Terminating Threads
			Passing Arguments to Threads
			Joining and Detaching Threads
			Stack Management
			Miscellaneous Routines
		Mutex Variables and Condition Variables
		Case Studies: Producer Consumer Problem

	Parallel Programming with Java
		Core Java
		Java for Parallel Programming
		Communication in Parallel Programming with Java
		PVM in Java
		MPI in Java
		OpenMP in Java

	MPI with Perl and Python
	GPGPU Technology
		Introduction to GPGPU
		GPU Performance Vs CPU Performance
		GPU Computing Application Domain
		nVIDIA's GPGPU Hardware Model
		GPU Computing with CUDA
			Thread Hierarchy
			Memory Hierarchy
			CUDA constructs
		Case-Studies (Algorithms and Parallelization Approaches)
			Matrix - Multiplication
			Cryptography Algorithms

	Parallel Programming on Heterogeneous Platforms with OpenCL
		Introduction to OpenCL
		The OpenCL programming language
		OpenCL Platform and Runtime API
		OpenCL Vs CUDA (Point interchangeable in places)
		Case-Studies *(Algorithms and Parallelization Approaches)

	Grid Computing and FPGA Based H/W Accelerators
		Grid Computing
			Introduction to Grid Computing
			Grid Middleware
			Grid Security
			Grid Applications
			Brief on algorithms and program development for Grids
		FPGA Based H/W accelerators
			Introduction to FPGAs & Cards
			Introduction to VHDL
			EDA tools
			Programming Style
	*Syllabus may be revised from time to time.

What can I do to prepare for the test?

The test measures how much comprehensive knowledge you have acquired over time through on-the-job experience and training. The following books will be an extremely valuable aide in preparing to take, and pass the test.

Recommended books for reference:

Reference Books for C, C++, FORTRAN, OS concepts, PERL, PYTHON, Parallel programming paradigm using MPI, OpenMP, GPU computing:

• C++: The Complete Reference by Herbert Schildt
• Distributed Operating Systems: Concepts and Design by P K Sinha
• Introduction to Parallel Computing by Ananth Grama, George Karypis, Vipin Kumar and Anshul Gupta
• Introduction to Programming with Fortran: With Coverage of Fortran 90,95, 2003, 2008 and 77 by Ian Chivers and Jane Sleightholme
• Learning Perl by Randal L. Schwartz, Brian D Foy and Tom Phoenix
• Operating System Concepts by Abraham Silberschatz, Peter B. Galvin and Greg Gagne
• Parallel Programming in C with MPI and OpenMP by Quinn, Michael J
• Parallel Programming in OpenMP by Rohit Chandra, Ramesh Menon, Leo Dagum and David Kohr
• Programming Massively Parallel Processors: A Hands-on Approach (Applications of GPU Computing Series) by David B. Kirk and Wen-mei W. Hwu
• Programming in ANSI C by Balaguruswamy
• Programming Python by Mark Lutz
• Using MPI and Using MPI-2 (Scientific and Engineering Computation) by William Gropp, Ewing L Lusk, Anthony Skjellum and Rajeev Thakur

Note: These books cover most of the test topics which will help in preparing for CCHPCP Certification exam. The above list is not comprehensive and should not be used as the only study material for this test.