OOTI 20 Year Anniversary Symposium
Abstracts
Cloud Computing
:Architecturing a
dynamic infrastructure
Cloud
Computing is an emerging computing model by which users can gain access to
their applications from anywhere, through any connected device. A
user-centric interface makes the cloud infrastructure supporting the
applications transparent to users. The applications reside in massively
scalable data centers where computational resources can be dynamically
provisioned and shared to achieve significant economies of scale. Thanks to
a strong service management platform, the management costs of adding more IT
resources to the cloud can be significantly lower than those associated with
alternate infrastructures.
What is driving the adoption of cloud computing? Many factors, including the
proliferation of smart mobile devices, high-speed connectivity, higher
density computing and data-intensive Web 2.0 applications. But also
cost-effectiveness and energy-efficacy.
As a result, vendors across the ICT industry have announced Cloud Computing
efforts of varying capabilities and among corporate clients there is an
increasing interest in aspects of the Cloud, such as infrastructure
outsourcing, software as a service key processes as a service and
next-generation distributed computing.
This presentation will dwell on the notion of Cloud Computing, with an
emphasis on the architecture needed to provide a request-driven, dynamic
allocation of computing resources for a mix of workloads on a massively
scalable, heterogeneous and virtualized environment. In addition, it will
gives examples how ICT companies, among them
LOFAR: A challenge in software architecture - by Ruud Overeem (ASTRON)
In the mid 90’s the idea was born to build a complete new
type of radio telescope, a telescope without dishes.
When you have enough (>10.000) dish-less antennas in the field it
must be possible to get a signal to noise ratio that is high enough for
doing science.
This idea opened the door to many new opportunities. Since you don't have
dishes anymore it must be possible to look in several directions
simultaneously! What's more, you are able to receive a very width band of
the radio spectrum instead of 1 or 2 relative small bands that is common for
conventional telescopes.
The telescope was named LOFAR (LOw Frequency ARray). Needless to say that
LOFAR brought a lot of challenges. How do you manage more than 10.000
antennas that are spread over an area with a diameter of 350 km? How do you
handle an incoming datastream of 14 Tb/s? How do you process 512 subbands
simultaneously in real time? How many TFlop do you need for processing the
datastreams?
After many years of research and hard work we finally found the solutions
for these (and a lot more) challenges. The presentation will show most of
the components of the LOFAR telescope and will drag you along in the
challenges we faced and decisions we made.
Not all mobile software is mobile
The
presentation will briefly touch on the architecture of a public mobile
communication network and then discuss the invocation of applications in an
IMS network on the Service Layer. The current Service Layer application
architecture will be described and finally a way to interface
applications on a mobile terminal to counterparts on the service layer.
Innovations in TV system
design - by Hans Rijns, NXP
Semiconductors
These challenges need to be faced within even more stringent constraints:
the fast growing environmental awareness makes people want their next TV
sets to consume far less electric power than today and of course be realized
at continously lowered price levels. This poses a wealth of innovation
challenges on set makers as well as on TV IC manufacturers.
In his presentation Hans Rijns will address exciting developments in three
areas of the TV system design. He will elaborate on the latest wideband
multi-channel receiver technology: a single tuner solution that combines
wideband signal reception with parallel digital channel decoding. It
provides full watching and recording flexibility, at lower cost levels than
offered by conventional single channel receivers. On the video processing
end he will show how low quality video input signals can be enhanced to high
quality motion pictures by the application of new smart algorithms. On the
display processing side, he will provide insight in the energy
characteristics of the image processing sequence for different types of
LCD-TV displays and explain how active backlight dimming technologies can
yield huge power savings, at even enhanced picture quality.
Asynchronous Silicon Compilation - by Ad Peeters (Handshake Solutions)
The design of integrated circuits is a continuous roller coaster in which
both the technology and the market are changing quickly. On the technology
side, Moore's law enables an exponential growth of the number of transistors
that can be integrated in silicon in ever smaller dimensions with ever
increasing design complexity. On the market side, windows of opportunities
are getting smaller and smaller, demanding ever faster design cycles.
In 1985, a project was started in Philips Research in which asynchronous
circuit design in combination with silicon compilation was proposed as a way
to efficiently design chips. This technology has lead to some successes,
most notably in the smartcard and identification domain, where several 100
millions of ICs have been sold.
Smart Dust: science, fiction or a real
business opportunity?
Rapid
advances in technology have enabled a new generation of tiny, inexpensive,
networked sensors. Sensors are tiny devices capable of capturing physical
information, such as heat, light or motion, about an environment. Embedding
millions of sensors into an environment creates a digital skin or wireless
network of sensors, each sensor capable of capturing physical information
about its immediate space. These massively distributed sensor networks
communicate with one another and summarize the immense amounts of low-level
information to produce data representative of the overall environment.
Collaborative, smart sensor networks present information in a qualitative,
human-interpretable form, which allows the system and the people to respond
intelligently.
The past several years of wireless sensor network research have resulted in
advancements in many areas. The generality and potential of this will be
investigated in various application domains. This talk will highlight some
of the challenges as we are advancing this technology from academic
prototypes to broad commercial usage.
