L
Lucas Roh
Researcher at Colorado State University
Publications - 16
Citations - 219
Lucas Roh is an academic researcher from Colorado State University. The author has contributed to research in topics: Dataflow & Code generation. The author has an hindex of 9, co-authored 16 publications receiving 219 citations. Previous affiliations of Lucas Roh include Argonne National Laboratory.
Papers
More filters
Proceedings ArticleDOI
An evaluation of bottom-up and top-down thread generation techniques
TL;DR: The authors study the memory referencing behavior of individual machine-level instructions using simulations of fully-associative caches under MIN replacement to obtain a deeper understanding of useful program behavior that can be eventually employed at optimizing programs and to motivate architectural features aimed at improving the efficacy of memory hierarchies.
Proceedings ArticleDOI
Design of storage hierarchy in multithreaded architectures
Lucas Roh,Walid Najjar +1 more
TL;DR: In this paper, a storage model which can exploit the locality of accesses to memory and its effect on the cache is developed and evaluated, and the results indicate there is a large amount of inter-thread locality that can be exploited and that we can get an efficient storage system by exploiting the characteristics of nonblocking threads.
Proceedings ArticleDOI
Algorithms and design for a second-order automatic differentiation module
TL;DR: This article describes approaches to computing second-order derivatives with automatic differentiation (AD) based on the forward mode and the propagation of univariate Taylor series and the underlying infrastructure used to create a language-independent translation tool.
Proceedings ArticleDOI
Generation and quantitative evaluation of dataflow clusters
TL;DR: The results indicate that even with a simple bottom-up algorithm for generating clusters, cluster execution offers a good speedup over the fine-grain execution over a wide range of architectures and coarse- grain execution is scalable, tolerates network latency and high matching cost well.
An Evaluation of Optimized Threaded Code Generation
TL;DR: This paper presents such bottom-up optimizations of multithreaded architectures as well as evaluating their eeectiveness in terms of overall performance and speciic thread characteristics such as size, length, instruction level parallelism, number of inputs and synchronization costs.