Identifying Data 2019/20
Subject (*) High Performance Architecture Code 614473101
Study programme
Mestrado Universitario en Computación de Altas Prestacións / High Performance Computing (Mod. Presencial)
Descriptors Cycle Period Year Type Credits
Official Master's Degree 1st four-month period
 First Obligatory 6
Language
English
Teaching method Face-to-face
Prerequisites
Department Departamento profesorado máster
Enxeñaría de Computadores
Coordinador
Doallo Biempica, Ramon
 E-mail
ramon.doallo@udc.es
Lecturers
Blanco Heras, Dora
Doallo Biempica, Ramon
Fernández Rivera, Francisco
Touriño Dominguez, Juan
Vázquez Álvarez, Álvaro
 E-mail
ramon.doallo@udc.es
juan.tourino@udc.es
Web http://aula.cesga.es
General description Mediante esta materia preténdese completar os coñecementos sobre arquitecturas de altas prestacións, para elo trátanse as arquitecturas paralelas modernas tanto desde o punto de vista funcional como de deseño. Estes coñecementos facilitarán ao alumno o desenvolvemento de algoritmos paralelos correctos e eficientes en base ás características arquitectónicas dos sistemas utilizados. As materias relacionadas coa programación vanse ver, por tanto, claramente beneficiadas.

Study programme competencies
Code Study programme competences
A1 CE1 - Define, evaluate and select the most appropriate architecture and software to solve a problem
A2 CE2 - Analyze and improve the performance of a given architecture or software
A3 CE3 - Know the high performance computing basic concepts
A4 CE4 - Deepen in the knowledge of different programming tools and programming languages in the field of the high performance computing
A8 CE8 - Be able to apply the acquired knowledge, capabilities and aptitudes to the profesional environment, planning, managing and evaluating project in the high performance computing field
B1 CB6 - Possess and understand the knowledge that give a baseline or opportunity to be original in the development and/or application of ideas, often in a research environment
B2 CB7 - The students have to know how to apply the acquired knowledge and their capacity to solve problems in new or hardly explored environment inside wider contexts (or multidiscipinary) related to its area of development
B3 CB8 - The students have to be able to integrate knowledge and face the complexity to make judgments from information, despite being partial and limited, includes reflexions about the social and ethical responsabilities linked to the application of their judgements and knowledge
B4 CB9 - The students have to be able to communicate their conclusions, their knowledge and the reasons that hold them to specialized and non specialized audience in a clear and unambiguous manner
B5 CB10 - The students have to possess learning skills that allows them to continue to study in a mainly self-driven or autonomous manner
B6 CG1 - Be able to search and select useful information to solve complex problems, using the bibliographic sources of the field
B7 CG2 - Elaborate adqueately and originally written essays or motivated reasonings, write planings, work projects, scientific papers and formulate reasonable hypothesis
B9 CG4 - Be able to plan and do research, development and innovation tasks in high performance computing related environments
B10 CG5 - Be able to work in teams, specially multidisciplinary, and do a proper time and people management and decision taking
C1 CT1 - Use the basic technologies of the information and computing technology field required for the professional development and the long-life learning

Learning aims
Learning outcomes Study programme competences
The student will know the different types of parallel architectures and their classification. AJ1
AJ3
 BJ1
BJ5
 CJ1
The student will study the basics about organization and design of a parallel architecture, both at microarchitecture level and multiprocessor systems level. AJ2
AJ8
 BJ2
BJ4
BJ6
The student will know the design principles an main componentes of a multiprocessor system. AJ2
AJ3
AJ8
 BJ1
BJ3
BJ7
BJ9
BJ10
 CJ1
The student will learn to analyse parallel architecture performance. AJ2
AJ4
AJ8
 BJ4
BJ7
BJ9
 CJ1

Contents
Topic Sub-topic
Chapter 1. Parallel computers - Historic introduction
- Levels of parallelism: form microarchitecture to supercomputers
- Classification
Chapter 2. Design of multiprocessors, multicores and manycores - Introduction
- Architecture of multiprocessors, multicores and manycores
- Memory architecture
Chapter 3. Cache Coherence
 - Protocols
- Snooping (UMA systems)
- Protocols based on directories (CC-NUMA systems)
Tema 4. Sincronización e consistencia de memoria en multiprocesadores - Primitivas de sincronización
- Soporte hardware para sincronización
- Implementaciones software de sincronización
- Modelos de consistencia de memoria
- Comparación entre os modelos de consistencia
Chapter 5. Interconexion networks - Types of networks
- Main components
- Performance
- Design
Chapter 6. Distributed systems: clusters - Introduction
- Cluster architecture
- Nodes
- Interconnection networks
- Software
- Tools
- Applications
- Load balance
Chapter 7. Introduction to performance analysis. - Motivation
- Basic concepts
- Characterization of performance issues
- Architecture features related to performance

Planning
Methodologies / tests Competencies Ordinary class hours Student’s personal work hours Total hours
Guest lecture / keynote speech A1 A3 B1 B5 22 0 22
Laboratory practice A2 A4 B2 B6 B10 C1 24 24 48
Supervised projects A8 B3 B4 B7 B9 0 72 72
Mixed objective/subjective test B4 B7 2 0 2
 
Personalized attention 6 0 6
 
(*)The information in the planning table is for guidance only and does not take into account the heterogeneity of the students.

Methodologies
Methodologies Description
Guest lecture / keynote speech The lecturer presents contents of the subject, and asks questions to the student in order to improve learning. There can also be discussions about specific topics.
Laboratory practice Practices and exercices are done in laboratory to support contents explained at keynote speech.
Supervised projects Students will develop individually or joined to other students specific projects/works. It could be possible to present to the rest to the students these works.
Mixed objective/subjective test Some questions about practice and supervised projects can be done by lecturer.

Personalized attention
Methodologies
Laboratory practice
Supervised projects
Description
Laboratory practice:
Lecturer and student analyse the practices done by the student.

Supervised projects:
Students receive lecturer guidance about their assigned supervised projects, and the acomplishment of the scheduled goals are verified periodically.

Assessment
Methodologies Competencies Description Qualification
Laboratory practice A2 A4 B2 B6 B10 C1 Valórase o correcto funcionamento, a estructuración do código, e aa comprensión dos conceptos traballados. Tamén valórase a participación activa do estudante durante as sesións de prácticas. 39
Supervised projects A8 B3 B4 B7 B9 No caso de desenvolvemento de código, valoranse os mesmos aspectos que nas prácticas. No caso de traballos escritos, valorase a capacidade de comprensión e síntesis sobre o tema proposto, e a calidade da presentación. 59
Mixed objective/subjective test B4 B7 Tanto no caso das prácticas como dos traballos tutelados o profesor pode facer preguntas concretas aos estudantes que poden complementar a avaliación. 2
 
Assessment comments

Evaluation is done in a continuous way based on the supervised projects delivered by the students (60%), and practices and active participation of the students (40%)

The student can be requested to identify themselves by an official identification document in the evaluation process.


 

Sources of information
Basic

Basic books:
1. Arquitectura de Computadores, Xullo Ortega, Mancia Anguita e Alberto Prieto. Thompson. 2005.
2. High Performance Cluster Computing, Rajkumar Buyya, ed., Prentice Hall PTR, 1999. ISBN 0-13-013784-7, 0-13-013785-5.
Complementary

Complementary books:
1. Parallel Computer Architecture, David E. Culler, Jaswinder Pal Singh e Anoop Gupta. Morgan Kaufmann Publishers. 1999.
2. In Search of Clusters, 2ª ed., Gregory Pfister, Prentice Hall, 1998, ISBN: 0138997090.
3. Organización e Arquitectura de Computadores (7ª edición), W. Stallings. Prentice Hall. 2007.
4. Computer Architecture: a Quantitative Approach (6ª edición), John L. Hennessy e David A. Patterson. Morgan Kaufmann Publishers. 2017.

Recommendations
Subjects that it is recommended to have taken before

Subjects that are recommended to be taken simultaneously
Parallel Programming/614473102

Subjects that continue the syllabus
Heterogeneous Programming/614473103
HPC on the Cloud/614473106
Advanced Parallel Programming/614473107

Other comments


(*)The teaching guide is the document in which the URV publishes the information about all its courses. It is a public document and cannot be modified. Only in exceptional cases can it be revised by the competent agent or duly revised so that it is in line with current legislation.