Summary: High Level Programming Methodologies for Data Intensive Computations
Gagan Agrawal 1
Renato Ferreira 2
Joel Saltz 3
Ruoming Jin 4
1 Introduction
Solving problems that have large computational and storage requirements is becoming increasingly critical for advances in
many domains of science and engineering. By allowing algorithms for such problems to be programmed in widely used or
rapidly emerging high­level paradigms, like object­oriented and declarative programming models, rapid prototyping and easy
development of computational techniques can be facilitated.
Our research focuses on an important class of scientific and engineering problems, data intensive computations. The
datasets associated with these applications are usually multi­dimensional. The data dimensions can be spatial coordinates,
time, or varying experimental conditions such as temperature, velocity or magnetic field. The increasing importance of such
applications has been widely recognized. Runtime systems like the Active Data Repository [5, 6] and the Passion runtime
library [18, 19] allow high performance on data intensive applications, but do not address the need for programming with
high­level abstractions.
We target two high­level programming models for this important class of computations:
1. Object­Oriented (Java Based): Object­oriented features like encapsulation and polymorphism can ease software devel­
opment and allow better long­term management of large computational projects. We target a dialect of Java for expressing
data intensive computations. This dialect of Java includes extensions for allowing multidimensional collection of objects,

Source: Agrawal, Gagan - Department of Computer Science and Engineering, Ohio State University

Collections: Computer Technologies and Information Sciences