|
Coordinators:
Pere Mato,
CERN |
|
This theme presents a selection of advanced underlying computing technologies which are particularly relevant in the context of scientific computing, and serve as a basis to construct higher level services services such as those offered by Grid Technologies. They include software engineering, computer architectures, computing security and networking topics
The first series of lectures presents modern techniques for software design and modern tools and technologies for understanding and improving existing software. The emphasis is placed on the large software projects and large executables that are common in HEP. The series consist of lectures and exercises. These lectures include topics such software engineering, design, methodology and testing.
The second series of lectures describes the evolution and the state of the art of computer architectures, discusses the bottlenecks and the consequences of this evolution on software design and optimization. It presents principles for writing software that scales with the hardware , techniques for hardware and software performance monitoring and issues related to the impact of compilers on performances.
The third topic addresses computer security with a particular focus on two aspects: cryptography, authentication and security infrastructures on the one hand, and the creation of secure software on the other hand. The latter series includes hand-on exercises.
The 4th topic addresses virtualization and cloud computing.
The theme is complemented by a series of lectures on networking, which presents principles, methods and techniques for improving quality of service and network performance. |
|
Series |
Type |
Lecture |
Description |
Lecturer |
|
|
|
|
|
|
|
Computer Architecture and Performance Tuning |
Lectures |
|||
|
Lecture 1 |
Understanding scalable hardware
|
|||
|
Lecture 2 |
Software that scales with the hardware
In
the second part of this double lecture we will discuss
several strategies which can allow software to scale to the
maximum resource potential in a given architecture. These
strategies are based on both data and task parallelism. We
will also mention the issue of “performance portability”
across platforms. Several important factors related to
programming styles and compiler invocation (flags, etc.)
will be reviewed. To back up everything with evidence,
several scalable examples from physics will be portrayed.
|
|||
|
Lecture 3 |
Understanding performance tuning
Performance tuning is an important step in application
development. Modern processor architectures often give us
the benefit of being able to look inside the application at
very low levels, however drawing high-level conclusions is
not always straightforward. The objective of this lecture is
to familiarize the attendees with the topic of performance
optimization and with certain common metrics which can be
used to define application efficiency. In addition, we will
demonstrate how to use language independent performance
tools in Linux, in order to obtain information about program
characteristics and bottlenecks.
|
|||
|
Exercises |
Exercise 1 Exercise 2 Exercise 3 |
The aim of the exercises in this series is to give the attendees a practical introduction to performance monitoring on Linux. Advanced tools will be used during the course, enabling the participants to discover how the structure of the code influences its performance. The participants will also be given the task of correlating performance figures with certain programming decisions. In addition, the participants will understand the limits of performance tuning and the ways to establish at which point inside those limits their workload is placed. The exercises will be supported by demonstrating real world problems in production environments, including multi-threaded examples. |
||
|
Prerequisite Knowledge |
Desirable prerequisite
and references to further information |
|
|
|
|
|
|
|
|
|
|
Networking QoS and Performance |
Lectures |
Lecture 1 |
Internet QoS options Improving Quality of Service guarantees and performances in data network is a key requirement of Grid computing. Indeed, fast transfers require high-bit rate connections, and grid operation requires network predictability and high availability. On the other hand, the Internet historical technology is not naturally best suited to deterministic behaviour. This lecture explains the technical challenges and the range of options available to improve QoS guarantees in Internet-based networks.
|
|
|
Lecture 2 |
Multimedia over the Internet The Grid is not only a network of computer resources but also a network of people cooperating to use these resources. Part of the collaborative tools scientists are increasingly using include audio and video systems. They place new challenging requirements on the networking systems. The class discusses these requirements and their consequences on the end-systems as well as within the underlying network.
|
|||
|
Prerequisite Knowledge
|
Mandatory prerequisite |
For this series of lectures, there is no mandatory pre-requisite knowledge, as long as the participants are professional computer scientists. |
|
|
|
Desirable prerequisite
and references to further information |
The participants will draw maximum benefits from the lectures if they have a fair knowledge of computer network principles, in particular the concepts of
|
|
||
|
Books
Vikipedia Computer Networking (http://en.wikipedia.org/wiki/Computer_networks) Other Links |
|
|||
|
Creating secure software
|
Lectures
|
Lecture 1 |
Introduction to computer security First lecture starts with a definition of computer security and an explanation of why it is so difficult to achieve. The lecture highlights the importance of proper threat modelling and risk assessment. It then presents three complementary methods of mitigating threats: protection, detection, reaction; and tries to prove that security through obscurity is not a good choice.
|
|
|
Lecture 2 |
Security in different phases of software development The second lecture addresses the following question: how to create secure software? It introduces the main security principles (like least-privilege, or defense-in-depth) and discusses security in different phases of the software development cycle. The emphasis is put on the implementation part: most common pitfalls and security bugs are listed, followed by advice on best practice for security development.
|
|||
|
Exercises |
Exercise 1 |
Avoiding, detecting and removing software security vulnerabilities
In
the practice session, a range of typical security
vulnerabilities will be presented. The goal is to learn how
they can be exploited (for privilege escalation, data
confidentiality compromise etc.), how to correct them, and
how to avoid them in the first place! Students will be given
small pieces of source code in different programming
languages, and will be asked to find vulnerabilities and fix
them. The online course documentation will gradually reveal
more and more information to help students in this task.
Additionally, students will have a chance to try several
source code analysis tools, and see how such tools can help
them find functionality bugs and security vulnerabilities.
|
||
|
Prerequisite Knowledge |
Desirable prerequisite
and
References to further information |
|
|
|
|
Books
|
|
|||
| Virtualisation |
Lectures |
Lecture 1 |
Introduction to virtualisation technology First lecture covers the definition and description of the various modes of virtualization techniques used in computing science, also from a historical perspective. Later we present the most recent advances and technology trends, in particular we will single out the server virtualization as a key enabling technology behind the emerging cloud computing paradigm.
|
Pere Mato |
|
Lecture 2 |
Applications of the virtualisation technology Second lecture deals with possible applications of server virtualization technology to support LHC computing effort. We will be using the CernVM project as an example to illustrate how a virtual machine can be crafted to act as an end-user work environment as well as job hosting environment running on cloud or Grid infrastructure.
|
Pere Mato | ||
|
Exercises |
Exercises 1 |
Using a virtual machine for data analysis The student will exercise how to setup a virtual machine and perform some simple analysis task (using ROOT for example) on a local host. Later we will show how the same task can be achieved using cloud resources such as Amazon EC2.
|
Predag Buncic Pere Mato |
|
|
Prerequisite Knowledge |
Mandatory prerequisite |
Be familiar with ROOT and very basic C++ and Linux/Unix environment. |
||
|
Desirable prerequisite |
||||
|
and References to further information |