Workshop in Computer Science (Fall 2002):
Workshop on Data Synopses
(0368.3500.19)
(classes will be specified separately below)
Announcement:
Meeting with teaching assistants: Wednesday, November 6, 11-13,
Schreiber 209.
Teaching Assistant:
Asaf Shapira (
asafico@tau.ac.il
),
Leon Portman (
leonpo@tau.ac.il
).
-
Introductory classes: Wednesday, October 16 & 23, 11-13, Schreiber 08.
-
Meeting with teaching assistant: Wednesday, November 6, 11-13, Schreiber 209.
-
General introduction to database systems and SQL by Alon Shalita, CS System's
DBA: Monday, November 18, 12-13. Schreiber 309.
- General introduction to Microsoft's .Net environment:
Monday, December 9: on your own at Microsoft's Tech-Ed Academic (Naftali Bldg)
One of the increasing challenges in many application domains is dealing
efficiently with massive data sets. Data repositories with multi-Gigabytes
and multi-Terabytes of data become common, and there are many applications
that need to deal with massive data sets, or with streaming data of huge
quantities, such as:
- Massive databases & data warehouses
- Telecom call records
- Financial applications
- Web logs and clickstreams
- Network monitoring and traffic engineering
- Network security & fraud detection
- Sensor networks
- Data mining
Traditional data structures are not suitable to handle massive data sets,
since available memory and performance constraints allows only for data
structures substantially smaller than such sets.
Synopsis data structures concisely capture appropriate characteristics of
large data sets, using relatively small memory, and allowing fast
computation of approximate queries.
Devising effective data synopses is a fast growing area of research.
This workshop will focus on advanced techniques for synopsis data
structures:
- Implement various synopsis data structures
- Test their performance
- Experience integration with an existing database platform (no prior
knowledge in databases is required)
- The workshop may be written in C++, C# or Java
- All projects will involve flavors of advanced algorithmics, programming, &
experimental work, with different levels of emphasis in the different projects.
Method: The various projects will be done in groups of up to three students.
The projects are aimed for an integrated working system, yet designed to be
done independently.
Please refer to
this page
for further details.
Seleceted projects could be nominated to the
projects competition
announced by the School of Computer Science.
matias+sadna02@cs.tau.ac.il
Last updated October, 2002