Vernon Austel

Learn More
We present a generic formal security model for operating systems of multiapplicative smart cards. The model formalizes the main security aspects of secrecy, integrity, secure communication between applications and secure downloading of new applications. The model satisfies a security policy consisting of authentication and intransitive noninterference. The(More)
First-principles simulations of high-Z metallic systems using the Qbox code on the BlueGene/L supercomputer demonstrate unprecedented performance and scaling for a quantum simulation code. Specifically designed to take advantage of massively-parallel systems like BlueGene/L, Qbox demonstrates excellent parallel efficiency and peak performance. A sustained(More)
This paper details our experiences with successfully validating a trusted device at FIPS 140-1 Level 4—earning the world’s first certificate at this highest level. Over the last several years, our group designed and built a physically secure PCI card (the IBM 4758 [5]) containing a general-purpose processor with crypto support. However, for this device to(More)
BLIS is a new software framework for instantiating high-performance BLAS-like dense linear algebra libraries. We demonstrate how BLIS acts as a productivity multiplier by using it to implement the level-3 BLAS on a variety of current architectures. The systems for which we demonstrate the framework include state-of-the-art general-purpose, low-power, and(More)
Secure coprocessors provide a foundation for many exciting electronic commerce applications, as previous work [20, 21] has demonstrated. As our recent work [6, 13, 14] has explored, building a high-end secure coprocessor that can be easily programmed and deployed by a wide range of third parties can be an important step toward realizing this promise. But(More)
We present two generic formal security models for operating systems of multiapplicative smart cards. The models formalize the main security aspects of secrecy, integrity, secure communication between applications and secure downloading of new applications. The first model is maximally abstract, whereas the second extends the first by adding practically(More)
BLIS is a new software framework for instantiating high-performance BLAS-like dense linear algebra libraries. We demonstrate how BLIS acts as a productivity multiplier by using it to implement the level-3 BLAS on a variety of current architectures. The systems for which we demonstrate the framework include state-of-the-art general purpose, low-power, and(More)
Deep Neural Networks (DNNs) have recently been shown to significantly outperform existing machine learning techniques in several pattern recognition tasks. DNNs are the state-of-the-art models used in image recognition, object detection, classification and tracking, and speech and language processing applications. The biggest drawback to DNNs has been the(More)
While Deep Neural Networks (DNNs) have achieved tremendous success for LVCSR tasks, training these networks is slow. To date, the most common approach to train DNNs is via stochastic gradient descent (SGD), serially on a single GPU machine. Serial training, coupled with the large number of training parameters and speech data set sizes, makes DNN training(More)
Existing commercial finite element analysis (FEA) codes do not exhibit the performance necessary for large scale analysis on parallel computer systems. In this paper, we demonstrate the performance characteristics of a commercial parallel structural analysis code, ADVC, on Blue Gene/L (BG/L). The numerical algorithm of ADVC is described, tuned, and(More)