All
lectures at iCSC2013
|
A few questions addressed
in the lecture |
|
|
|
Monday
25 February |
|
|
10:15 11:10 |
Lecture
1 |
Introduction
to parallel computing on GPUs |
Felice
Pantaleo
|
|
Description
All computing systems, from mobile
to supercomputers, are becoming heterogeneous parallel
computers using both multi-core CPUs and many-thread GPUs
for higher power efficiency and computation throughput.
While the computing community is racing to build tools and
libraries to ease the use of these heterogeneous parallel
computing systems, effective and confident use of these
systems will always require knowledge about the low-level
programming interfaces in these systems.
This lecture is
designed to introduce through examples, based on the CUDA
programming language, the three abstractions that make the
foundations of GPU programming:
- Thread hierarchy
- Synchronization
- Memory hierarchy/Shared Memory
The aim of this lecture is to give the audience a solid
foundation on which to start building their own first GPU
application.
|
|
Audience and benefits
This lecture targets physicists and engineers who are
interested in improving the performance of their software by
using off-the-shelf graphic car.
After this lecture, the attendees are expected to have a
good understanding of the principles that govern parallel
programming in CUDA and will be able to write their first
GPU-accelerated application. |
|
Pre-requisite
Listeners don't need to have advanced knowledge about
parallel computing or GPU programming
|
|
A few questions addressed
in the lecture |
|
|
|
Monday
25 February |
|
|
11:30 12:25 |
Lecture
2 |
Use of GPUs for triggering in HEP experiments |
Felice
Pantaleo
|
|
Description
This lecture describes a pilot
project for the use of GPUs in a real-time triggering
application in the early trigger stages at the
NA62
experiment at CERN, and the results of the first field tests
together with a prototype data acquisition (DAQ) system.
This pilot project within NA62 aims at integrating GPUs
into the central L0 trigger processor, and also to use them
as fast online processors for computing trigger primitives.
The online use of GPUs would allow the computation of more
complex trigger primitives already at this first trigger
level.
|
|
Audience and benefits
This lecture targets physicists and engineers who are
involved in the design of future triggering systems.
After this lecture, the attendees are expected to have a
good understanding of the reason why GPUs are considered a
key technology for the development of a "software-based"
triggering system.
|
|
Pre-requisite
knowledge and dependencies
Participants having not attended
lecture 1 may have difficulties to fully follow this
lecture, if they are not somewhat familiar with GPU
programming principles and abstractions.
|
|
A few questions addressed
in the lecture |
|
|
|
Monday
25 February |
|
|
14:00 14:55 |
Lecture
3 |
Introduction to the human visual system and image
pre-processing |
Samuele Carli
Martin Hellmich
|
|
Description
This lecture gives an introduction
to the problem of computer vision starting with a comparison
of the human visual system to its artificial approximations.
It will give an overview of different example
applications.
Later on it focuses on image
acquisition processing techniques which are needed for
higher level processing: basic entities representations,
filters, transformations, image equalization and camera
models.
|
|
Audience and benefits
This lecture targets everyone interested in the basic
concepts behind vision, both human and artificial, as well
as in basic image manipulation, filtering, enhancement and
representation.
After this lecture the attendees are
expected to have acquired a basic knowledge of the important
concepts, properties and limitations of image acquisition
systems and basic image manipulation techniques, and will be
able to evaluate and compare different aspects of human and
artificial vision systems.
|
|
Pre-requisite
This lecture can be followed by
anyone having interest in the subject and basic knowledge in
linear algebra and mathematics.
Nevertheless the main
concepts explained in this lecture will be understandable by
everyone.
|
|
A few questions addressed
in the lecture |
|
|
|
Monday
25 February |
|
|
14:00 14:55 |
Lecture
4 |
Introduction to image feature detection and 3D
reconstruction |
Samuele Carli
Martin Hellmich
|
|
Description
This lecture introduces the
concepts needed to obtain depth information from two or more
images taken from slightly different angles.
It will
cover the various steps necessary, from features detection
and matching to triangulation techniques.
|
|
Audience and benefits
This lecture targets everyone interested in the basic
concepts behind computer vision, how 3D cameras work and
what are the necessary steps in order to reconstruction
depth information.
After this lecture the attendees
are able to understand the concepts behind 3D cameras (and
movies) and are able to use this knowledge to mechanically
reconstruct a scene as a cloud of 3D points.
They will be
able as well to understand how stereo vision helps
self-driving cars to estimate distances and avoid
collisions. |
|
Pre-requisite
This lecture can be followed by
anyone having interest in the subject and with basic
knowledge of linear algebra.
Lecture 1 of this lectures series is
recommended if you are not familiar with the concepts of
image manipulation and vision systems.
Nevertheless the
main concepts explained in this lecture will be
understandable by everyone.
|
|
A few questions addressed
in the lecture |
|
|
|
Monday
25 February |
|
|
16:30
17:25 |
Lecture
5 |
Introduction to object recognition and scene
understanding |
Samuele Carli
Martin Hellmich
|
|
Description
This lecture gives an introduction
to object recognition. It expands the topic of stereo vision
and describes different object recognition techniques,
involving object discrimination through texture and
geometric properties (like shape, occlusion and overlapping
of objects) to perform both instance recognition and
classification.
It gives a soft comparison between
artificial recognition and how the human brain recognizes
items in its environment.
|
|
Audience and benefits
This lecture targets everyone interested in the topic of
artificial object recognitors and classifiers, as well as
techniques for texture and geometric properties
discrimination.
After this lecture the attendees are
able to understand the main challenges which computer vision
is facing as well as some of the basic ideas and musings
behind existing and future solutions.
|
|
Pre-requisite
This lecture can be followed by
anyone having interest in the subject and with basic
knowledge of linear algebra. The preceding lectures of this
series are recommended if you are not familiar with the
concepts of image manipulation and vision systems or image's
features detection and 3D reconstruction.
Nevertheless
the main concepts explained in this lecture will be
understandable by everyone..
|
|
A few questions addressed
in the lecture |
|
|
|
Monday
25 February |
|
|
16:30
17:25 |
Lecture
5 |
Introduction to object recognition and scene
understanding |
Samuele Carli
Martin Hellmich
|
|
Description
This lecture gives an introduction
to object recognition. It expands the topic of stereo vision
and describes different object recognition techniques,
involving object discrimination through texture and
geometric properties (like shape, occlusion and overlapping
of objects) to perform both instance recognition and
classification.
It gives a soft comparison between
artificial recognition and how the human brain recognizes
items in its environment.
|
|
Audience and benefits
This lecture targets everyone interested in the topic of
artificial object recognitors and classifiers, as well as
techniques for texture and geometric properties
discrimination.
After this lecture the attendees are
able to understand the main challenges which computer vision
is facing as well as some of the basic ideas and musings
behind existing and future solutions.
|
|
Pre-requisite
This lecture can be followed by
anyone having interest in the subject and with basic
knowledge of linear algebra. The preceding lectures of this
series are recommended if you are not familiar with the
concepts of image manipulation and vision systems or image's
features detection and 3D reconstruction.
Nevertheless
the main concepts explained in this lecture will be
understandable by everyone..
|
|
A few questions addressed
in the lecture |
|
|
|
Tuesday
26 February |
|
|
09:30 10:30 |
Lecture
6 |
Testing for development, deployment
and operation |
Ramon Medrano Llamas
|
|
Description
This lectures is comprised of two
parts:
- tools and techniques for development and
deployment
- tools and methods for large scale
operation.
-
In the first part,
new methodologies and tools that have arisen in the past
few years not only to help the development but also to
smoothly deploy features on live systems automatically
driven by test results will be presented and compared.
Keywords: SCM, agile
methodologies, TDD, continuous integration, cloud
deployments, legacy management, deployment rollback.
-
In the second part,
modern methodologies and tools focused on operation will
be presented. Based on live testing, auto recovery and
proactive monitoring, they help reducing the fire
fighting overhead and increasing the autonomy of the
system.
Keywords: DevOps,
live testing, system wide profiling, autonomic
computing, monitoring.
|
|
Audience
This theme is aimed at software developers interested in
automating the quality assurance process of their
software development projects and/or get a sneak peek on
latest testing tools.
It also targets those
involved in operation of computing systems willing to
improve and expedite their response to incidents and
requests.
Benefits
The attendees are expected to gain an overview of the
testing techniques, tools and motivations during the
development process and why testing automation is essential.
They will also get an overview of the new operational
techniques based on improved relationships between
development and operation teams.
|
|
Pre-requisite
knowledge and dependencies
Little is required, but knowledge of
common software tool chains, classic software development
strategies and common software life will help fully benefit
from the lecture.
|
|
A few questions addressed
in the lecture |
|
|
|
Tuesday
26 February |
|
|
11:00
12:00 |
Lecture
7 |
How the LHC experiment have interpreted the Grid
distributed computing model |
Mattia
Cinquilli
|
|
Description
Yet another talk on Grids?
You
may think you know all you need to know about Grid and its usage
by LHC experiments. But do you know that every experiment has
his own interpretation of the grid model? Why did they make
different interpretation? Was it because of different
requirements, by chance?
The major differences lye in data
management and workload management. What does it mean in
practice? But Grids are not set in stone, they evolve, so
what next. After a comparative analysis of the Grid
interpretations and implementations by LHC experiments, the
lecture will conclude with discuss future trends and
possible evolution scenarios.
|
|
Audience and benefits
This lecture targets computer scientists and physicists
interested in an overview of the Grid paradigm explained
through a comparison of its implementation by LHC's
experiments.
At the end of the lecture, participants, in particular
computer scientists, are expected to have a better
understanding of how the specific services they use or are
working on fit in the global Grid picture.
|
|
Pre-requisite
This lecture can be reasonably
followed by anyone having some
minimum knowledge of distributed computing.
|
|