Open Access. Powered by Scholars. Published by Universities.®
Physical Sciences and Mathematics Commons™
Open Access. Powered by Scholars. Published by Universities.®
Articles 1 - 4 of 4
Full-Text Articles in Physical Sciences and Mathematics
On Modeling Contention For Shared Caches In Multi-Core Processors With Techniques From Ecology, Amy Apon, Wesley Emeneker
On Modeling Contention For Shared Caches In Multi-Core Processors With Techniques From Ecology, Amy Apon, Wesley Emeneker
Amy W. Apon
Multi-core x86_64 processors introduced an important change in architecture, a shared last level cache. Historically, each processor has had access to a large private cache that seamlessly and transparently (to end users) interfaced with main memory. Previously, processes or threads only had to compete for memory bandwidth, but now they are competing for actual space. Competition for space and environmental resources is a problem studied in other scientific domains. This paper introduces methods from ecology to model multi-core cache usage with the competitive Lotka–Volterra equations. A model is presented and validated for characterizing the interaction of cores through shared caching, …
A Component-Based End-To-End Simulation Of The Linux File System, Amy Apon, Hai Nguyen
A Component-Based End-To-End Simulation Of The Linux File System, Amy Apon, Hai Nguyen
Amy W. Apon
The Linux file system is designed with components utilizing a layered architecture. The upper components hide details of the lower components, and each layer presents unified and simple interfaces to the layers above and below. This design helps Linux to be flexible as well as to provide support for multiple types of storage devices. In this paper, this component architecture is used to develop a realistic simulation without having to model lower level details of the hardware layer or particular storage devices. A detailed simulation-based performance model of the Linux ext3 file system has been developed using Colored Petri Nets. …
Community Funding Models For Computational Resources, Amy Apon, Jeff Pummill, Dana Brunson
Community Funding Models For Computational Resources, Amy Apon, Jeff Pummill, Dana Brunson
Amy W. Apon
As scientific research has extended far beyond the practicality and abilities of laboratory experiments, computational simulations have become the mainstay of enabling and furthering the research in a way never previously thought possible. It is becoming commonplace to model and simulate both the very large, such as black hole collisions in astrophysics, and the very small, such as subatomic particle behavior and interaction in high energy physics. In addition to the previous examples detailing extremes, practically every area of research currently utilizes and benefits from computational resources to simulate their work; financial modeling, weather forecasting, geological phenomena, geo-spatial data analysis, …
Jummp: Job Uninterrupted Maneuverable Mapreduce Platform, William Moody, Linh Ngo, Edward Duffy, Amy Apon
Jummp: Job Uninterrupted Maneuverable Mapreduce Platform, William Moody, Linh Ngo, Edward Duffy, Amy Apon
Amy W. Apon
In this paper, we present JUMMP, the Job Uninterrupted Maneuverable MapReduce Platform, an automated scheduling platform that provides a customized Hadoop environment within a batch-scheduled cluster environment. JUMMP enables an interactive pseudo-persistent MapReduce platform within the existing administrative structure of an academic high performance computing center by “jumping” between nodes with minimal administrative effort. Jumping is implemented by the synchronization of stopping and starting daemon processes on different nodes in the cluster. Our experimental evaluation shows that JUMMP can be as efficient as a persistent Hadoop cluster on dedicated computing resources, depending on the jump time. Additionally, we show that …