None

We introduce a new Condensed History algorithm for the Monte Carlo simulation of electron transport. To obtain more accurate simulations, the new algorithm preserves the mean position and the variance in the mean position exactly for electrons that have traveled a given path length and are traveling in a given direction. This is accomplished by deriving the zeroth-, first-, and second-order spatial moments of the Spencer-Lewis equation and employing this information directly in the Condensed History process. Numerical calculations demonstrate the advantages of our method over standard Condensed History methods.

Current trends in high performance computing suggest that users will soon have widespread access to clusters of multiprocessors with hundreds, if not thousands, of processors. This unprecedented degree of parallelism will undoubtedly expose scalability limitations in existing applications, where scalability is the ability of a parallel algorithm on a parallel architecture to effectively utilize an increasing number of processors. Users will need precise and automated techniques for detecting the cause of limited scalability. This paper addresses this dilemma. First, we argue that users face numerous challenges in understanding application scalability: managing substantial amounts of experiment data, extracting useful trends from this data, and reconciling performance information with their application's design. Second, we propose a solution to automate this data analysis problem by applying fundamental statistical techniques to scalability experiment data. Finally, we evaluate our operational prototype on several applications, and show that statistical techniques offer an effective strategy for assessing application scalability. In particular, we find that non-parametric correlation of the number of tasks to the ratio of the time for individual communication operations to overall communication time provides a reliable measure for identifying communication operations that scale poorly.

The results of five decades of experimental investigations of open-ended and closed magnetic confinement geometries are examined to see if intrinsic topology-dependent differences in their cross-field transport can be discerned. The evidence strongly supports a picture in which closed systems (stellarators, tokamaks, reversed-field pinches, etc.) are in all cases studied to date characterized by some level of plasma turbulence, leading to substantial deviations from purely classical cross-field transport. This transport is often describable as a Bohm-like scaling with plasma temperature and magnetic field intensity. By contrast, open systems have in many significant examples been able to approach closely to classically predicted cross-field transport, including cases where the transport appeared to be more than five orders of magnitude slower than the Bohm-diffusion rate. To explain these differences the following tentative hypothesis is put forward: The differences arise from two sources: (1) differences in the instability driving terms arising from free-energy sources, such as current flow along the field lines, etc. and, (2) differences in the nature of the boundary conditions for the various unstable waves that may be stimulated by these free energy sources within the plasma. By analogy with a laser, closed systems, with their flux tubes returning on themselves, …

This is one of two companion papers that describe the ENERGY-10 PV design tool computer simulation program. The other paper is titled ''ENERGY-10 Photovoltaics: A New Capability.'' Whereas this paper focuses on the PV aspects of the program, the companion paper focuses on the implementation method. The case study in this paper is a commercial building application, whereas the case study in the companion paper is a residential application with an entirely different building load characteristic. Together they provide a balanced view.