KTU 2024 Scheme · Semester 5 · Program Elective II — Computer Science and Engineering
Advanced Computer Architecture (PECST528) Syllabus
Official KTU 2024 Scheme syllabus for Advanced Computer Architecture, Semester 5, Program Elective II — Computer Science and Engineering (Computer Science and Engineering).
Course Code
PECST528
Credits
3
Teaching Hours
3:0:0:0 (L:T:P:R)
CIE Marks
40
ESE Marks
60
Exam Duration
2 Hrs 30 Min
Prerequisites
PBCST404
Semester
Semester 5
Course Objective
To introduce advanced processor architectures including parallelism concepts in programming of multiprocessors and multicomputers, and to provide a detailed understanding of data flow in computer architectures.
Module-wise Syllabus
Module 1
Introduction: the impact of hardware and software technology trends. Self review - instruction set architecture, memory addressing, addressing modes, classes of computers, concept of computer hardware and organization. Measuring, reporting and summarizing performance, benchmarks - desktop and server, Amdahl's law, the processor performance equation. Beyond the books: explore the high performance computing benchmarks at spec.org and compare results submitted by different vendors.
Module 2
Review of the basic concepts of parallel processing and pipelining. Instruction level parallelism, data dependencies and hazards, different types of dependences, compiler techniques for ILP, branch prediction - correlating branch predictor. Dynamic scheduling - idea, introduction to Tomasulo's scheme, register renaming, hardware speculation, reorder buffers. Multiple issue and static scheduling, VLIW.
Module 3
Data level parallelism: vector processors - how they work, memory banks, stride, scatter-gather; SIMD comparison with vector and GPU, comparison of loops in C vs CUDA, NVIDIA GPU memory structure, vector processor vs GPU, multimedia SIMD computers vs GPU. Multiprocessor architecture - centralized shared memory architecture, cache coherence and snooping protocol (implementation details not required), performance of symmetric shared-memory processors, distributed shared memory and directory-based protocol (basics), synchronization - basic hardware primitives, memory consistency models - sequential and relaxed.
Module 4
Warehouse scale computers: goals and requirements, programming frameworks for batch processing - MapReduce and Hadoop, computer architecture of warehouse-scale computers. Moore's law, Dennard scaling, dark silicon and the transition towards heterogeneous architectures. Asymmetric multi-core architecture - static and dynamic. Functional heterogeneous multicore architecture - GPUs, accelerators, reconfigurable computing. Beyond the textbook: identify the processor used in your PC and mobile phone and study its architecture.
Course Outcomes
- CO1Enumerate the different classes of computers and where they are used in everyday life.
- CO2Compute the effect of hardware/software enhancements on the speedup of a processor using Amdahl's law.
- CO3Interpret possible dependencies that can cause hazards in a given block of code.
- CO4Summarize different strategies followed to ensure instruction level parallelism.
- CO5Compare different strategies followed to ensure instruction level parallelism and different strategies followed to ensure data parallelism.
- CO6Illustrate the need for memory consistency models and cache coherence protocols and explain the principle behind them.
Assessment Pattern (CIE: 40 marks, ESE: 60 marks)
Continuous Internal Evaluation (CIE)
| Attendance | 5 |
| Assignment / Microproject | 15 |
| Internal Examination 1 (Written) | 10 |
| Internal Examination 2 (Written) | 10 |
End Semester Examination (ESE)
Total 60 marks, 2 Hrs 30 Min. See the official KTU syllabus document for the exact Part A / Part B question pattern for this course.
Textbooks & Reference Books
Textbooks
- Computer Architecture: A Quantitative Approach — John L. Hennessy, David A. Patterson (Morgan Kaufmann, 5th edition, 2012)
- The Dark Side of Silicon: Energy Efficient Computing in the Dark Silicon Era — Anil Kanduri et al. (Springer, 1st edition, 2017)
Reference Books
- Computer Architecture — Gérard Blanchet, Bertrand Dupouy (Wiley, 1st edition, 2013)
- Advanced Computer Architectures — Sajjan G. Shiva (Taylor & Francis, 1st edition, 2018)
- Computer Architecture — Charles Fox (No Starch Press, 1st edition, 2024)
Frequently Asked Questions
How many credits is KTU Advanced Computer Architecture (PECST528)?
3 credits, with 3:0:0:0 (L:T:P:R) teaching hours per week, under the KTU 2024 Scheme.
How many modules are in the PECST528 syllabus?
4 modules.
What is the CIE and ESE mark split for this course?
CIE (Continuous Internal Evaluation): 40 marks. ESE (End Semester Examination): 60 marks, 2 Hrs 30 Min. Total: 100 marks.
What are the recommended textbooks for PECST528?
Computer Architecture: A Quantitative Approach (John L. Hennessy, David A. Patterson); The Dark Side of Silicon: Energy Efficient Computing in the Dark Silicon Era (Anil Kanduri et al.).
Is this syllabus specific to one branch, or common to others too?
This Semester 5 course is listed under Program Elective II — Computer Science and Engineering at KTU under the 2024 Scheme — check the course header above for which branches it's common to.
Need help with KTU Advanced Computer Architecture?
Learnizo offers live, 1-on-1 online tuition for KTU CSE subjects — matched to your exact module and semester.