Link to
Colorado State University Home Page

BMAC Fall 2006: Abstracts


The Evolution of FreeBSD and Linux Clemente Izurieta
Department of Computer Science
Colorado State University

Is the nature of Open Source Software (OSS) evolution fundamentally different from that of the traditional and commercially available software systems? Lehman and others conducted a series of empirical studies that found that traditional systems grow at a linear or sub-linear rate. A prior case study of the Linux OSS system suggests that OSS may evolve in a unique manner. Godfrey and Tu found that some aspects of Linux are growing at a super-linear rate rather than a sub-linear rate. Additional studies are necessary before drawing conclusions. Thus, we examine the evolution of FreeBSD and re-analyze the evolution of Linux, and find evidence that the growth of both systems has a linear upper bound, and thus appear to grow at similar rates to that of commercial systems. These results do not support the hypothesis that OSS systems grow at rates that exceed that of traditional systems.

BIO

Clem Izurieta is a graduate student at Colorado State University. Born in Chile, Clem has a bachelor of Mathematics from the University of Wollongong, NSW, Australia, and a Masters degree in Computer Science from Montana State University. Clem has worked in industry for many years including 12 years at the Unix Research and Development Laboratories of Hewlett Packard and most recently 2 years at the Fort Collins Design Center of Intel Corporation. Clem's interests include algorithms, building software systems, and all aspects of software engineering associated with this.


Mixed Discrete and Continuous Algorithms for Scheduling Airborne Astronomy Observations
Jeremy Frank
Planning and Scheduling Group
NASA Ames Research Center

We describe the problem of scheduling astronomy observations for the Stratospheric Observatory for Infrared Astronomy, an airborne telescope. The problem requires maximizing the number of requested observations scheduled subject to a mixture of discrete and continuous constraints relating the feasibility of an astronomical observation to the position and time at which the observation begins, telescope elevation limits, Special Use Airspace limitations, and available fuel. Solving the problem requires making discrete choices (e.g. selection and sequencing of observations) and continuous ones (e.g. takeoff time and setup actions for observations by repositioning the aircraft). Previously, we developed an incomplete algorithm called ForwardPlanner using a combination of AI and OR techniques including progression planning, lookahead heuristics, stochastic sampling and numerical optimization, to solve a simplified version of this problem. While initial results were promising, ForwardPlanner fails to scale when accounting for all relevant constraints. We describe a novel combination of Squeaky Wheel Optimization (SWO), an incomplete algorithm designed to solve scheduling problems, with previously devised numerical optimization methods and stochastic sampling approaches, as well as heuristics based on reformulations of the SFPP to traditional OR scheduling problems. We show that this new algorithm finds as good or better flight plans as the previous approaches, often with less computation time.
Paper:
J. Frank and E. Kurklu
Mixed Discrete and Continuous Algorithms for Scheduling Airborne Astronomy Observations, Proceedings of the 2nd International Conference on Constraint Programming, Artificial Intelligence and Operations Research, 2005


Viewing HIV as Evolutionary Computation
Dr. Richard K. Belew
University of California at San Diego

Anyway you measure, the viral quasi-species HIV is a prodigious evolutionary engine. It produces on the order of 1010 virions daily, and the reverse transcription process exploited by HIV creates extremely high mutation rates. While anti-retroviral drug therapies now are often successful against wildtype virus, many drug-resistant mutants have also evolved and can be transmitted to newly infected individuals. Currently 14,000 new infections occur daily, with 95% of these occurring in the developing world least able to afford the drugs that have been developed.

The scale of the epidemic has also made HIV one of the most well-studied biological systems. We will take this opportunity to connect several key features of this BIOLOGICAL system to standard features of evolutionary COMPUTATIONS (EC). This suggests several unusual extensions to standard EC, both towards better models of biology and as effective algorithms.


Vulnerabilities in Browsers and Servers
Omar Alhazmi and Sung-Whan Woo
Colorado State University

In this talk, we present some recent results of software security assessment research. We examine the feasibility of quantitatively characterizing the vulnerabilities in the two major HTTP servers. In particular, we investigate the applicability of quantitative empirical models to the vulnerabilities discovery process for these servers. Such models can allow us to predict the number of vulnerabilities that may potentially be present in a server but may not yet have been found. The data on vulnerabilities found in the two servers is mined and analyzed. We explore the applicability of a time-based and an effort-based vulnerability discovery model. The effort-based model requires the use of the current market-share of a server. Both models have been successfully used for vulnerabilities in the major operating systems. Our results show that both vulnerabilities discovery models fit the data for the HTTP servers well. We also examine two separate classification schemes for server vulnerabilities, one based on the source of error and the other based on severity, and then explore the applicability of the quantitative methods to individual classes.

Browsers represent one of the most used software systems. Since they serve as the gateway to the web, vulnerabilities in browsers can have great impact. Recently there has been considerable debate about the vulnerabilities in the two major browsers Microsoft Internet Explorer and Mozilla Firefox which represent two opposite development paradigms. Here we present a quantitative perspective involving vulnerability detection rates, severity, and patch development. The available data suggests that the popular perceptions can sometimes be inaccurate and a detailed quantitative analysis of the data is needed for a careful evaluation of the risk. Making projections for the near future requires an understanding of the longer term trends. The need for reconciling alternating conventions for enumerating the vulnerabilities is also identified.

Categorized vulnerabilities analysis has shown that some vulnerabilities categories are more severe than others, and some vulnerabilities categories are usually less severe. This can be used as a guideline to design better test cases that give a higher priority to special categories in order to optimize the testing and reduce the cost of testing.