Gyrokinetic particle-in-cell optimization on emerging multi- and manycore platforms

The next decade of high-performance computing (HPC) systems will see a rapid evolution and divergence of multi- and manycore architectures as power and cooling constraints limit increases in microprocessor clock speeds. Understanding efficient optimization methodologies on diverse multicore designs in the context of demanding numerical methods is one of the greatest challenges faced today by the HPC community. In this work, we examine the efficient multicore optimization of GTC, a petascale gyrokinetic toroidal fusion code for studying plasma microturbulence in tokamak devices. For GTC’s key computational components (charge deposition and particle push), we explore efficient parallelization strategies across a broad range of emerging multicore designs, including the recently-released Intel Nehalem-EX, the AMD Opteron Istanbul, and the highly multithreaded Sun UltraSparc T2+. We also present the first study on tuning gyrokinetic particle-in-cell (PIC) algorithms for graphics processors, using the NVIDIA C2050 (Fermi). Our work discusses several novel optimization approaches for gyrokinetic PIC, including mixed-precision computation, particle binning and decomposition strategies, grid replication, SIMDized atomic floating-point operations, and effective GPU texture memory utilization. Overall, we achieve significant performance improvements of 1.3-4.7× on these complex PIC kernels, despite the inherent challenges of data dependency and locality. Our work also points to several architectural and programming features that could significantly enhance PIC performance and productivity on next-generation architectures.     

The role of individual and system characteristics in computerized training systems

Two studies examined the role of characteristics that are important to consider in the development of computerized training systems. The first study employed covariance structure modeling and tested a hypothesized model regarding the relationship among three latent variables and a set of measured variables serving as indicators for the latent variables. The overall model fit quite well according to both statistical and practical assessments. Results indicate that, for naive users, a latent variable reflecting an individual's attitude toward computers has a significant influence on the other two latent variables, one being how the computerized task is perceived, and the second being the choice of interacting with a computer in the future. A structural path representing the influence of the current experience with the computer has a small directional influence on future interaction preference with a person rather than a computer. A second study examined system characteristics and focused specifically on preferences for how much power an expert system should be given. Scenarios were developed depicting three different types of expert systems, each with three different levels of power. Results indicate that individuals prefer those systems with either low or moderate power, those which do not initiate action without human intervention. Systems that initiate action on their own are not viewed positively. The discussion focuses on implications for future training systems.     

Optimal replacement time of an equipment via simulation for truncated failure distributions

The optimal replacement time T^* of an equipment for different failure (truncated) distributions is obtained by Monte Carlo simulation and the results verified by analytical methods, whenever possible. The results are illustrated by explicit numerical solutions (wherever feasible) and by simulation runs in all cases.     

Wireless Internet Access Using IS-2000 Third Generation System: A Performance and Capacity Study

The Internet has been growing tremendously in the recent years and applications like web browsing are becoming increasingly popular. In a collective effort to provide seamless access to the Internet, wireless equipment manufacturers and service providers are developing 3G wireless systems that efficiently support current and future Internet applications. In this paper, we evaluate the performance and capacity of a 3G wireless data system based on IS-2000 standard. We consider web browsing as the common application for all users and evaluate the system performance for single and parallel web browsing sessions. We perform this study through a detailed simulation of web traffic model described by distributions of number of objects per page, object size, page request size and page reading time. The simulation includes HTTP and TCP/IP protocols, link level recovery, radio resource management, mobility, channel model and, delays in the Internet and the radio access network. We quantify important system attributes like average page download times and system throughput (Kb/s per carrier per sector). We also evaluate normalized object download time, normalized page download time, penalty in performance due to link errors, link layer buffer sizes needed, channel holding time, average power used and distribution of the power used in the system.     

Detecting the Origin of Cancer‐Mobile Quantum Probe for Single Cancer Stem Cell Detection

Cancer stem cells (CSC) are believed to be the driving force of cancer metastases and are a rare population of self‐renewing cells that contribute majorly to the poor outcomes of cancer therapy. The detection of CSC is of utmost importance to shed light on the indestructible nature of certain solid tumors and their metastatic ability. However, tumors tend to harbor CSCs in a specialized niche, making the detection process difficult. Currently, there is no method available to detect CSCs. The significance of this work is twofold. First, to the best of the knowledge, it is the first time that the detection of CSC is demonstrated. This approach simultaneously detects both the phenotypic and the metabolic state of the cell, thus enabling universal detection of CSC with high accuracy. Second, to the best of the knowledge, for the first time, light is shed on cell chemistry of CSC in their dedicated niche to facilitate a better understanding of the key players involved in the metabolic rewiring of CSC. This work will enable a better understanding of the fundamentals of CSCs, which are critical for the early diagnosis of cancer and the development of therapies for the cure of cancer.     

Enhancing web browser security against malware extensions

In this paper we examine security issues of functionality extension mechanisms supported by web browsers. Extensions (or “plug-ins”) in modern web browsers enjoy unrestrained access at all times and thus are attractive vectors for malware. To solidify the claim, we take on the role of malware writers looking to assume control of a user’s browser space. We have taken advantage of the lack of security mechanisms for browser extensions and implemented a malware application for the popular Firefox web browser, which we call browserSpy, that requires no special privileges to be installed. browserSpy takes complete control of the user’s browser space, can observe all activity performed through the browser and is undetectable. We then adopt the role of defenders to discuss defense strategies against such malware. Our primary contribution is a mechanism that uses code integrity checking techniques to control the extension installation and loading process. We describe two implementations of this mechanism: a drop-in solution that employs JavaScript and a faster, in-browser solution that makes uses of the browser’s native cryptography implementation. We also discuss techniques for runtime monitoring of extension behavior to provide a foundation for defending threats posed by installed extensions.