MSc in Parallel and Distributed Computer Systems

University Politehnica of Bucharest

Program Description

MSc in Parallel and Distributed Computer Systems

University Politehnica of Bucharest

Advanced Techniques in Computer and Network Security

The course is focused on current issues in computer and network security. There are discussed security issues of the electronic information from the perspective of ensuring the confidentiality, integrity, authenticity, and non-repudiation of the information. The course is a natural continuation of the discipline of “Cryptography” and addresses from practical perspective of the cryptographic techniques and protocols (authentication, secret sharing, key exchange, subliminal channels), with applications in e-mail security, IP security, Web security and security of computer network (protected by firewall, IDS, IPS devices and antivirus protection techniques).

Also, the course presents principles used in security evaluation of cryptographic modules, products, and applications derived from industrial security standards: Security Requirements for Cryptographic Modules (ISO 19790) and Common Criteria for Information Technology Security Evaluation (ISO 15408).

Cluster and Grid Computing

The course covers the specific concepts of Cluster and GRID computing. It is presented the basic concepts of the cluster, as a new approach of parallel and distributed processing system, which consists of a collection of interconnected stand-alone heterogeneous systems cooperatively working together as a single, integrated computing resource. In the course are presented the type of clusters, cluster architecture, new concept in OS services for distributed processing, physical cluster interconnections and interconnect support, cluster programming environments, monitoring and performance analysis tools.

The course presents the essence of Grids how to utilize highly flexible network architectures, and how to sharing of all computing resources, not just data. Are presented the grid technologies, an extensible and open Grid architecture, general aspects of basic components that enable interoperability among different Grid resources. There are presented the principal Grid characteristics: Wide geographical distribution, Heterogeneous, Resource sharing, Multiple admin policies, Resource coordination, Transparent access, Dependable, Consistent, Pervasive. It is presented a Sample Grid Computing Environment: Resource Sharing & Aggregation and Grid Architecture for Computational Economy. The Layered Grid Architecture is presented. In the second part of the course, the students have to choose a topic of application of the Grid computing and they have to present an essay about it.

Computer and Network Security

Computer and Network Security class covers the topic of computer and network security, from advanced cryptographic techniques, innovative algorithms, security protocols and policies, trust ensuring infrastructures and specific technologies for securing networks. At the end of this class, the students should be able to: develop a complex security policy for ensuring the safety of the provided services, identify threats and vulnerabilities at network and operating system level, identify types of attacks and mitigate them, identify the vulnerabilities of layer 2 and 3 equipments, implement a solution for ensuring a high level of security for the equipments, monitor the network, servers and workstations for identifying and stopping attacks, configuring a firewall, an IDS/IPS and an AAA system.

This class also has an important practical component, which includes: configuring routers that define the network perimeter with the use of specific instruction set for ensuring a high level of security, configuring layer 2 and 3 equipments in a safe manner, configuring a firewall for allowing basic network operations, configuring an IDS/IPS, configuring AAA both on routers and switches.

Distributed algorithms

Distributed algorithms are a very actual and important area in Computer Science. This is a master level class, in which we assume the students are already familiar with the fundamental notions and models of distributed algorithms.

First, we will discuss the causal order and total order. We will present the problems and discuss why traditional solutions will not work in all cases. We will then discuss the possible solutions, both in academia and industry and study some of the most important research papers that appeared on these topics. A particular example of a distributed application, a collaborative text editor, will be discussed and we will see how we can relax the strong conditions from the total order for this particular case by introducing the operational transformation technique. Finally, we will discuss distributed consensus and peer to peer systems.

Distributed Systems

The course aims at teaching and integrating the main concepts, principles, models, and techniques for the development of distributed system software. Another objective is increasing students’ capacity to use the acquired knowledge in modeling, the design of software components for distributed systems, implementing middleware programs based on modern technologies, evaluating the implemented systems against specification and design criteria, and optimizing the performance of solutions with engineering methods and instruments. Course Content: Models and architectures of large-scale dynamic distributed systems. P2P systems for content distribution: structured and unstructured architectures, search methods; content replication techniques; anonymity and reputation management.

Event-driven distributed systems: architecture, components; complex event processing, ECA (event, condition, action) model and finite-state automata with interval timestamps; intelligent engines for event processing. Cloud computing: resource provisioning, efficient use of resources; energy management; data security; data storage and management techniques. Compute and data-intensive models, MapReduce. Context-based distributed systems: context sensing, communication, and processing; ontology-based context models; security and privacy.

Parallel Programming

The Parallel Programming lecture is outlining a series of programming paradigms in the context of modern parallel computer architectures.

First, an overview of parallel programming models is given considering various programming models, issues such as productivity, performance, and portability and presenting a number of models for communication, synchronization, memory consistency and runtime systems. In this context, in the second part of the lecture, a series of parallel programming paradigms with shared (OpenMP, PThreads, Cilk, TBB, HPF, Chapel, Fortress, Stapl), and distributed memory (MPI, Charm++, Stapl), parallel global address shared space (UPC, X10) as well as other atypical paradigms are presented (Linda, MapReduce, MATLAB DCE). A basic understanding of computer architectures, operating systems, parallel and distributed algorithms and compilers is assumed.

This school offers programs in:
  • English


Last updated April 6, 2018
Duration & Price
This course is Campus based
Start Date
Start date
Sept. 2019
Duration
Duration
Price
Price
Information
Deadline
Locations
Romania - Bucharest, Bucharest
Start date : Sept. 2019
Application deadline Request Info
End date Request Info
Dates
Sept. 2019
Romania - Bucharest, Bucharest
Application deadline Request Info
End date Request Info