Computer Systems Architecture

The course is not on the list Without time-table
Code Completion Credits Range
14YAPS KZ 3 2+1
Department of Applied Informatics in Transportation

The content of this course is based on the description of computer architecture principles, e.g. von Neumann's model, pipelining, different coupling of system, transmission protocols, systolic systems, parallel systems classification, etc. Software levels of parallel systems are described, the basic method of synchronization are explained on the examples of CONVEX C3800, Cray Y-MP, Hitachi, Connection Machine - 5, IBM 3090 VF etc.<br>\r\nThe Amdahl's law, effectiveness of parallel systems performance measurement, RISC and CICS architectures, superscalar architectures, mass parallel architectures and symmetric multiprocessing are also explained on the examples IBM SP2, HP/Convex SPP1200, NEC SX-4 etc.<br>\r\nContents:<br>\r\n1.Basic types of computer architectures, von Neumann's model, microprogramming control.<br>\r\n2.Pipelining, array of processors, tightly and loosely coupled systems, systolic systems, data-flow architectures, associative memory, parallel systems classification.<br>\r\n3.Software level of parallelism, granularization, mutual exclusion synchronization, binary and general semaphores, vectorization and parallelization.<br>\r\n4.Classification of network topology, different coupling methods of processors and memory, examples of transmission protocols. <br>\r\n5.Supercomputers and minisupercomputers, vector and super-scalar architecture, mass parallel systems - examples.<br>\r\n6.Amdahl's law, effectiveness of parallel systems performance measurement, pear and real performance benchmarking, LINPACK, <br> parametrization of effectiveness.<br>\r\n7.Transputers, architecture, instruction set, T800 a T9000, memory organization, virtual channels, programming.<br>\r\n8.Principles of parallel systems languages, OCCAM, principles of HPF, Parallel Computing Forum<br>\r\n9.Architectures CICS and RISC, principles of design, basic features, advantages and disadvantages. Principles of superpipelining<br>\r\n10.Introduction to the neural networks, relation to the biological systems, perceptron, Rosenblat's rule, formal model of neuron. <br>\r\n11.Neuron models with different types of non-linearities, Radial Base Function (RBF).<br>\r\n12.Different models of neural networks, Hopfield's model, Kohonen's networks, back propagation learning.<br>\r\n13.Utilization of neural networks, classification and prediction, application in medicine, machinery, chemistry, etc. Design of multiplayer neural networks.

Syllabus of lectures:
Syllabus of tutorials:
Study Objective:
Study materials:
Further information:
No time-table has been prepared for this course
The course is a part of the following study plans:
Data valid to 2020-08-04
For updated information see http://bilakniha.cvut.cz/en/predmet23358105.html