Joshua W. Knight

Fractal nature of software-cache interaction

This paper uses fractals to model the clustering of cache misses. The clustering of cache misses can be quantified by a single number analog to a fractional dimension, and we are intrigued by the possibility that this number can be used as a measure of software complexity. The essential intuition is that cache misses are a direct reflection of changes in locality of reference, and that complex software requires more frequent (and larger) changes in this locality than simple software. The cluster dimension provides a measure (and perhaps the basis for a model) of the intrinsic differences between workloads. In this paper, we focus on cache miss activity as a discriminate between interactive and batch environments.

Simulation And Analysis Of A Pipeline Processor

In this paper we describe a software simulator (a timer) that is used to model a wide range of pipeline processors. A set of performance equations is developed that allow a user to separate the performance of a processor into its infinite-cache and finite-cache performance values. We then use the timer to study the performance of two different machine organizations. Performance curves are presented that help a user compare the performance of each organization (in terms of MIPS and cycles per instruction) to the cycle time chosen to implement the design.

Contrasting instruction-fetch time and instruction-decode time branch prediction mechanisms: Achieving synergy through their cooperative operation

Abstract We present a new design in which two branch prediction mechanisms are used in conjunction. We show that the combination of these mechanisms will reduce branch penalty, while also reducing chip area.

Stratification of amyotrophic lateral sclerosis patients: a crowdsourcing approach

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease with substantial heterogeneity in clinical presentation with an urgent need for better stratification tools for clinical development and care. In this study we used a crowdsourcing approach to address the problem of ALS patient stratification. The DREAM Prize4Life ALS Stratification Challenge was a crowdsourcing initiative using data from >10,000 patients from completed ALS clinical trials and 1479 patients from community-based patient registers. Challenge participants used machine learning and clustering techniques to predict ALS progression and survival. By developing new approaches, the best performing teams were able to predict disease outcomes better than currently available methods. At the same time, the integration of clustering components across methods led to the emergence of distinct consensus clusters, separating patients into four consistent groups, each with its unique predictors for classification. This analysis reveals for the first time the potential of a crowdsourcing approach to uncover covert patient sub-populations, and to accelerate disease understanding and therapeutic development.