First Astro-Grid Workshop
Computer Science, Queen's University Belfast
Monday 29 and Tuesday 30 January 2001
Preparations for the computational grid in the UK are accelerating.
Astronomy has important network- and compute-related needs, which are
quantitative and qualitative. The objective of this workshop is to
table these issues for discussion. We will draw a balance-sheet in
regard to how far we have come. Most importantly, we will sketch out
some of the most pressing near-future problems and issues.
The main themes of this workshop are the following.
- Data mining.
The computational grid will be used for the development of innovative
data mining methods. (Keywords: self-organization, multiple
resolution, progressive refinement.) Closely associated with this
work will be the development of new tools for ingest of data to data
archives and data centres.
- Information Discovery.
The grid will help the astronomer face the tsunamis of data resulting
from current and near-future observing facilities. Sub-themes:
dealing with structured and unstructured data, search and discovery
tools, smart information agents.
Observed data and in addition simulated data (from cosmological
models, and instrument and detector models) will be used.
Approaches will include interactive visual user interfaces, 3D
representations, and innovative new techniques aiming at support
for distributed, collaborative work.
- "The future of memory" - protocols for the preservation of data.
- On the fly recalibration and real-time pipelines.
- Spatial and temporal resolution.
- The Earth observation approach (analyse data and throw it away).
- The value of data, and anomaly/serendipitous investigation.
- "The computer is the network" but is I/O and in particular data
capture capability keeping pace? Some aspects of I/O, e.g.
knowledge elicitation, are very hard.
- Visualization and usability engineering.
- What does human perception really want? Really need?
- Can the benefits of scientific visualization be quantified?
(In education, in communication, in serendipitous discovery?)
- The Grid@Home - lessons from
Folding@home - from genome to
structure, "the peoples' grids".
- Crucial near future needs in data handling, processing and analysis.
Where and how will grid-enabled solutions be applied?
- Facilitating knowledge discovery - what are the necessary and
desirable characteristics of future data archive designs? How will
future data archives differ from present-day data archives?
Logistics and Travel Information
Starts Monday morning 09:00, January 29, 2001, dinner Monday evening,
finishes lunchtime Tuesday, January 30, 2001.
Day 1: Monday, January 29: Presentations. OHP will be available as
will a PC attached to a data projector with MS Powerpoint and a PDF
reader, and with a CD drive. For any other requirements, please check
Day 2: Tuesday, January 30: Round-table discussions on the above themes.
Travel information, maps, accommodation (Bed and Breakfast, other),
restaurants and cafes, weather, etc.: see under "Visitor Information"
Venue: Monday January 29, 2001, School of Computer Science, Queen's
University Belfast, Bernard Crossland Building, 18 Malone Road.
Room 1.15 (first floor, to left).
It is hoped to have at least all sessions on Monday January 29 webcast
live using streaming video, and available on-line afterwards.
Venue Tuesday January 30, 2001 (to be confirmed): Wellington Park Hotel
(across the street from the Bernard Crossland Building).
Last update: 2000-12-4. Contact:
F. Murtagh, email@example.com