Computer and Information Science Faculty Publications

TOP-C: a task-oriented parallel C interface

Gene D. Cooperman — Fri, 21 Jan 2011 08:25:23 PST

The goal of this work is to simplify parallel application development, and thus ease the learning barriers faced by non-experts. It is especially useful where there is little data-parallelism to be recognized by a compiler. The applications programmer need learn the intricacies of only one primary subroutine in order to get the full benefits of the parallel interface. The applications programmer defines a high level concept, the task, that depends only on his application, and not on any particular parallel library. The task is defined by its three phases: (a) the task input, (b) sequential code to execute the task,...

Parallelization of Geant4 using TOP-C and Marshalgen

Gene D. Cooperman — Wed, 22 Dec 2010 13:56:05 PST

Geant4 is a very large, highly accurate toolkit for Monte Carlo simulation of particle-matter interaction. It has been applied to high-energy physics, cosmic ray modeling, radiation shields, radiation therapy, mine detection, and other areas. Geant4 is being used to help design some high energy physics experiments (notably CMS and Atlas) to be run on the future large hadron collider: the largest particle collider in the world. The parallelization, ParGeant4, represents a challenge due to the unique characteristics of Geant4: (i) complex object-oriented design; (ii) intrinsic use of templates and abstract classes to be instantiated later by the end user; (iii)...

Transparent adaptive library-based checkpointing for master-worker style parallelism

Gene D. Cooperman — Wed, 22 Dec 2010 13:56:04 PST

We present a transparent, system-level checkpointing solution for master-worker parallelism that automatically adapts, upon restart, to the number of processor nodes available. This is important, since nodes in a cluster fail. It also allows one to adapt to using multiple cluster partitions and multiple resources from the Computational Grid, as they become available. Checkpointing a master-worker computation has the additional advantage of needing to checkpoint only the master process. This is both fast and more economical of disk space. This has been demonstrated by checkpointing Geant4, a million line C++ program. Our solution has been implemented in the context of...

Corrections to enhanced optical nonlinearity of superlattices

Gene D. Cooperman — Fri, 10 Dec 2010 13:05:47 PST

In recent publications, a large enhancement of the third order nonlinear optical susceptibility was predicted for GaAs-GaAIAs superlattices, as a result of the band nonparabolicities introduced by the additional periodicity of the superlattice. These predictions, based on the tight binding model, are here extended to the more realistic Kronig-Penney model. Results show that corrections to tight binding are non-negligible; however, enhancements of χ⁽³⁾ are still large, but reduced by approximately 30%-50% over previous estimates.

...

Self-pulsing and chaos in distributed feedback bistable optical devices

Herbert G. Winful — Fri, 10 Dec 2010 13:05:46 PST

We show that the light transmitted by a nonlinear distributed feedback structure can be steady (time independent), periodic, or chaotic depending on the intensity of the input cw beam. The feasibility of an experimental demonstration of such behavior is discussed.

A new current-voltage relation for duct precipitators valid for low and high current densities

Gene D. Cooperman — Fri, 10 Dec 2010 13:05:43 PST

A closed-form analytic current-voltage formula for duct electrostatic precipitators is presented. A short discussion of previous theoretical and numerical solutions is given, followed by an explanation of the theoretical formula derived here. A comparison with experimental data is then given, showing that the present formula is accurate over a wide range of conditions, including wide plate spacing.

Adaptive checkpointing for master-worker style parallelism (extended abstract)

Gene D. Cooperman — Fri, 10 Dec 2010 13:05:39 PST

DMTCP: transparent checkpointing for cluster computations and the desktop

Jason Ansel — Fri, 10 Dec 2010 13:05:37 PST

DMTCP (Distributed MultiThreaded CheckPointing) is a transparent user-level checkpointing package for distributed applications. Checkpointing and restart is demonstrated for a wide range of over 20 well known applications, including MATLAB, Python, TightVNC, MPICH2, OpenMPI, and runCMS. RunCMS runs as a 680 MB image in memory that includes 540 dynamic libraries, and is used for the CMS experiment of the Large Hadron Collider at CERN. DMTCP transparently checkpoints general cluster computations consisting of many nodes, processes, and threads; as well as typical desktop applications. On 128 distributed cores (32 nodes), checkpoint and restart times are typically 2 seconds, with negligible run-time...

Fast query processing by distributing an index over CPU caches

Xiaoqin Ma — Fri, 10 Dec 2010 13:05:35 PST

Data intensive applications on clusters often require requests quickly be sent to the node managing the desired data. In many applications, one must look through a sorted tree structure to determine the responsible node for accessing or storing the data. Examples include object tracking in sensor networks, packet routing over the internet, request processing in publish-subscribe middleware, and query processing in database systems. When the tree structure is larger than the CPU cache, the standard implementation potentially incurs many cache misses for each lookup; one cache miss at each successive level of the tree. As the CPURAM gap grows, this...

Geant4 developments and applications

J. Allison — Fri, 10 Dec 2010 13:05:31 PST

Geant4 is a software toolkit for the simulation of the passage of particles through matter. It is used by a large number of experiments and projects in a variety of application domains, including high energy physics, astrophysics and space science, medical physics and radiation protection. Its functionality and modeling capabilities continue to be extended, while its performance is enhanced. An overview of recent developments in diverse areas of the toolkit is presented. These include performance optimization for complex setups; improvements for the propagation in fields; new options for event biasing; and additions and improvements in geometry, physics processes and interactive...

General bounds on statistical query learning and PAC learning with noise via hypothesis boosting

Javed A. Aslam — Thu, 04 Nov 2010 08:58:54 PDT

We derive general bounds on the complexity of learning in the Statistical Query model and in the PAC model with classification noise. We do so by considering the problem of boosting the accuracy of weak learning algorithms which fall within the Statistical Query model. This new model was introduced by Kearns to provide a general framework for efficient PAC learning in the presence of classification noise. We first show a general scheme for boosting the accuracy of weak SQ learning algorithms, proving that weak SQ learning is equivalent to strong SQ learning. The boosting is efficient and is used to...

The Kerf toolkit for intrusion analysis

Javed A. Aslam — Thu, 04 Nov 2010 08:58:53 PDT

To aid system administrators with post-attack intrusion analysis, the Kerf toolkit provides an integrated front end and powerful correlation and data-representation tools, all in one package.

Semi-supervised data organization for interactive anomaly analysis

Javed A. Aslam — Thu, 04 Nov 2010 08:58:52 PDT

We consider the problem of interactive iterative analysis of datasets that consist of a large number of records represented as feature vectors. The record set is known to contain a number of anomalous records that the analyst desires to locate and describe in a short and comprehensive manner. The nature of the anomaly is not known in advance (in particular, it is not known, which features or feature values identify the anomalous records, and which are irrelevant to the search), and becomes clear only in the process of analysis, as the description of the target subset is gradually refined. This...

Distributed energy-conserving routing protocols

Qun Li — Wed, 03 Nov 2010 15:02:38 PDT

This paper discusses several distributed power-aware routing protocols in wireless ad-hoc networks (especially sensor networks). We seek to optimize the lifetime of the network. We have developed three distributed power-aware algorithms and analyzed their efficiency in terms of the number of message broadcasts and the overall network lifetime modeled as the time to the first message that can not be sent. These are: (1) a distributed min Power algorithm (modeled on a distributed version of Dijkstra's algorithm), (2) a distributed max-min algorithm, and (3) the distributed version of our the centralized online max-min zP_min algorithm presented in [12]. The first...

Management-based regulation: prescribing private management to achieve public goals

Cary Coglianese — Mon, 01 Nov 2010 14:27:57 PDT

We analyze a little-studied regulatory approach that we call "management-based" regulation. Management-based regulation directs regulated organizations to engage in a planning process that aims toward the achievement of public goals, offering firms flexibility in how they achieve public goals. In this paper we develop a framework for assessing conditions for using management-based regulation as opposed to the more traditional technology-based or performance-based regulation. Drawing on case studies of management-based regulation in the areas of food safety, industrial safety, and environmental protection, we show how management-based regulation can be an effective strategy when regulated entities are heterogeneous and regulatory outputs are...

Information and contact-making in policy networks: a model with evidence from the U.S. health policy domain

Daniel Carpenter — Mon, 01 Nov 2010 14:27:56 PDT

Theory: The political information that lobbyists seek is distributed in a communications network. Individual lobbyists must therefore choose their contacts carefully. We wed rational choice theory to network analysis in a combinatorial optimization model of lobbyists’ choice of contacts in a network. The model demonstrates the growing importance of political "friends" relative to acquaintances as contacts when the competition for information among groups rises. Hypotheses: Our model predicts that when the general demand for political information is low, a cocktail equilibrium prevails: lobbyists will invest their time in gaining "weak tie" political acquaintances rather than in gaining "strong tie" political...

The network structure of exploration and exploitation

David Lazer — Mon, 01 Nov 2010 14:27:55 PDT

Whether as team members brainstorming or cultures experimenting with new technologies, problem solvers communicate and share ideas. This paper examines how the structure of communication networks among actors can affect system-level performance. We present an agent-based computer simulation model of information sharing in which the less successful emulate the more successful. Results suggest that when agents are dealing with a complex problem, the more efficient the network at disseminating information, the better the short-run but the lower the long-run performance of the system. The dynamic underlying this result is that an inefficient network maintains diversity in the system and is...