David W Denning, Andrew Amos, Peter Rudge et al. · 1972
chronic arthritis.5 The virus divides prefer-entially in rapidly dividing cells particularly erythroid precursors, but by the time the rash appears, viraemia is no longer detectable. This suggests that the neuralgic amyotrophy seen in this patient is due to an idiosyncratic immune response. We recommend serological testing for parvovirus IgM in cases of neuralgic amy-otrophy to ascertain the relative causal con-tribution to this painful and disabling condition.
Cited by 0
Shachi Bhatt, Raul Diaz, Paul A. Trainor et al. · 1997
Neural crest cells (NCCs) comprise a multipotent, migratory cell population that generates a diverse array of cell and tissue types during vertebrate development. These include cartilage and bone, tendons, and connective tissue, as well as neurons, glia, melanocytes, and endo-crine and adipose cells; this remarkable lineage potential persists into adult life. Taken togeth-er with a limited capacity for self-renewal, neural crest cells bear the hallmarks of stem and progenitor cells and are considered to be synonymous with vertebrate evolution. The neural crest has provided a system for exploring the mechanisms that govern developmental process-es such as morphogenetic induction, cell migration, and fate determination. Today, much of the focus on neural crest cells revolves around their stem cell-like characteristics and potential for use in regenerative medicine. A thorough understanding of the signals and switches that govern mammalian neural crest patterning is central to potential therapeutic application of these cells and better appreciation of the role that neural crest cells play in vertebrate evolution,
Cited by 0
Carlo Fantozzi, Andrea Pietracaprina, Geppino Pucci · Journal of Parallel and Distributed Computing · 2005
Cited by 6
Anne M. Rogers · 1990
Parallel computers provide a large degree of computational power for programmers who are willing and able to harness it. The introduction of high-level languages and good compilers made possible the wide use of sequential machines but the lack of such tools for parallel machines hinders their widespread acceptance and use. Programmers must address issues such as process decomposition, synchronization, and load balancing. This is a severe burden and opens the door to time-dependent bugs, such as race conditions between reads and writes which are extremely difficult to detect. In this thesis, we use compile--time analysis and automatic restructuring of programs to exploit a two--level memory hierarchy. Many multiprocessor architectures can be modelled as two-level memory hierarchies, including message-passing machines such as the Intel iPSC/2. We show that such an approach can exploit data locality while avoiding the overhead associated with run-time coherence management. At the same time, it relieves the programmer from the burden of managing process decomposition and synchronization by automatically performing these tasks. We have developed a parallelizing compiler which, given a sequential program and a memory layout of its data, performs process decomposition while balancing parallelism against locality of reference. A process decomposition is obtained by specializing the program, for each processor, to the data that resides on that processor. If this analysis fails, the compiler falls back to a simple but inefficient scheme called run-time resolution. Each process''s role in the computation is determined by examining the data required for execution at run-time. Thus, our approach to process decomposition is ``data-driven'''' rather than ``program-driven.'''' We discuss several message optimizations that address the issues of overhead and synchronization in message transmission. Accumulation reorganizes the computation of a commutative and associative operator to reduce message traffic. Pipelining sends a value as close to its computation as possible to increase parallelism. Vectorization of messages combines messages with the same source and the same destination to reduce overhead. Our results from experiments in parallelizing SIMPLE, a large hydrodynamics benchmark, for the Intel iPSC/2, show a speed-up within sixty to seventy percent of hand-written code.
Cited by 24
Susanne Albers, Lene M. Favrholdt, Oliver Giel · Journal of Computer and System Sciences · 2004
Cited by 66
Susanne Albers, Lene M. Favrholdt, Oliver Giel · 2002
Motivated by the fact that competitive analysis yields too pessimistic results when applied to the paging problem, there has been considerable research interest in refining competitive analysis and in developing alternative models for studying online paging. The goal is to devise models in which theoretical results capture phenomena observed in practice.In this paper we propose a new, simple model for studying paging with locality of reference. The model is closely related to Denning's working set concept and directly reflects the amount of locality that request sequences exhibit. We demonstrate that our model is reasonable from a practical point of view.We use the page fault rate to evaluate the quality of paging algorithms, which is the performance measure used in practice. We develop tight or nearly tight bounds on the fault rates achieved by popular paging algorithms such as LRU, FIFO, deterministic Marking strategies and LFD. It shows that LRU is an optimal online algorithm, whereas FIFO and Marking strategies are not optimal in general. We present an experimental study comparing the page fault rates proven in our analyses to the page fault rates observed in practice. This is the first such study for an alternative/refined paging model.
Cited by 17
Susanne Albers, Lene M. Favrholdt, Oliver Giel · 2002
Motivated by the fact that competitive analysis yields too pessimistic results when applied to the paging problem, there has been considerable research interest in refining competitive analysis and in developing alternative models for studying online paging. The goal is to devise models in which theoretical results capture phenomena observed in practice.In this paper we propose a new, simple model for studying paging with locality of reference. The model is closely related to Denning's working set concept and directly reflects the amount of locality that request sequences exhibit. We demonstrate that our model is reasonable from a practical point of view.We use the page fault rate to evaluate the quality of paging algorithms, which is the performance measure used in practice. We develop tight or nearly tight bounds on the fault rates achieved by popular paging algorithms such as LRU, FIFO, deterministic Marking strategies and LFD. It shows that LRU is an optimal online algorithm, whereas FIFO and Marking strategies are not optimal in general. We present an experimental study comparing the page fault rates proven in our analyses to the page fault rates observed in practice. This is the first such study for an alternative/refined paging model.
Cited by 5
Amos Fiat, Manor Mendel · 1998
In this paper we refine the locality of reference concept captured by the access graph model to allow temporal changes in the behavior of the underlying process. We formalize this by introducing the concept of an extended access graph. We consider a graph parameter that captures the degree of change rate allowed. We study this new model and give truly online algorithms that are strongly competitive for extended access graphs. We can do so for almost all values of for which it is possible.
Cited by 0
Xuzhen He (14676114) · Figshare · 2023
<p>(a) Locality for 1D problems on the CPU platform. (b) Locality for 1D problems on the GPU platform. (c) Locality for 2D problems on the CPU platform. (d) Locality for 2D problems on the GPU platform.</p>
Cited by 0Open access
Allan Borodin, Sandy Irani, Prabhakar Raghavan et al. · Journal of Computer and System Sciences · 1995
Cited by 161
Spyros Angelopoulos, Reza Dorrigiv, Alejandro López-Ortíz · Lecture notes in computer science · 2008
Cited by 32
Allan Borodin, Prabhakar Raghavan, Sandy Irani et al. · 1991
while more practical than worst-case analysis. With these definitions, Sleator and Tarjan showed that no deterministic on-line paging algorithm can achieve a competitiveness less than k, and that a number of algorithms used in practice (including Least Recently Used or LRU and First-In First-Out or FIFO) are kcompetitive and thus optimal by this measure.
Cited by 85Open access
M. Kobayashi · 1980
Cited by 3
Wilfried N. Gansterer, Dieter F. Kvasnicka, Christoph W. Ueberhuber · 1997
This paper shows how the symmetric eigenproblem, which is the computationally most demanding part of numerous scientific and industrial applications, can be solved much more efficiently than by using algorithms currently implemented in Lapack routines. The main techniques used in the algorithm presented in this paper are (i) sophisticated blocking in the tridiagonalization, which leads to a two-sweep algorithm; and (ii) the computation of the eigenvectors of a band matrix instead of a tridiagonal matrix. This new algorithm improves the locality of data references and leads to a significant improvement in the floating-point performance of symmetric eigensolvers on modern computer systems. Speedup factors of up to four (depending on the computer architecture and the matrix size) have been observed. Keywords: Numerical Linear Algebra, Symmetric Eigenproblem, Tridiagonalization, Performance Oriented Numerical Algorithm, Blocked Algorithm Contents 1 Introduction 3 2 Eigensolver with Improv...
Cited by 0
Susanne Albers, Sonja Lauer · Journal of Computer and System Sciences · 2016
Cited by 5Open access
Ernst J. Haunschmid, Christoph W. Ueberhuber · 1997
This paper presents the principle ideas involved in hierarchical blocking, introduces the block packed storage scheme, and gives the implementation details and the performance rates of the hierarchically blocked Cholesky factorization. In some cases the newly developed routines are faster by an order of magnitude than the corresponding Lapack routines. Introduction Most current computers based on advanced architectures have hierarchical memories in which the data access in the upper levels of the memory hierarchy (registers, caches, and/or local memory) is faster than in the lower levels (shared or off-processor memory). One technique for exploiting the power of such machines more efficiently is the development of algorithms that maximize the reuse of data in the upper levels of memory. This can be done by partitioning matrices into blocks and by performing the computation using matrix-matrix or at least matrix-vector operations on the blocks. A set of basic linear algebra subroutines...
Cited by 0
Reza Dorrigiv, Alejandro López-Ortíz · Information Processing Letters · 2012
Cited by 5
2008
Abstract Motivated by the fact that competitive analysis yields too pessimistic re-sults when applied to the paging problem, there has been considerable research interest in refining competitive analysis and in developing alternativemodels for studying online paging. The goal is to devise models in which theoretical results capture phenomena observed in practice.In this paper we propose a new, simple model for studying paging with locality of reference. The model is closely related to Denning&apos;s working setconcept and directly reflects the amount of locality that request sequences exhibit. We demonstrate that our model is reasonable from a practical pointof view. We use the page fault rate to evaluate the quality of paging algorithms,which is the performance measure used in practice. We develop tight or nearly tight bounds on the fault rates achieved by popular paging algorithmssuch as LRU, FIFO, deterministic Marking strategies and LFD. It shows that LRU is an optimal online algorithm, whereas FIFO and Marking strategiesare not optimal in general. We present an experimental study comparing the page fault rates proven in our analyses to the page fault rates observed inpractice. This is the first such study for an alternative/refined paging model.
Cited by 0
Susanne Albers, Dario Frascaria · Algorithmica · 2018
Cited by 1Open access
Rodrigo L. C. Fonseca, Advisors Prof, Ion Stoica et al. · 2003
under the guidance of Prof. Ion Stoica and Prof. Scott Shenker, as a member of the Networking Group and of the RAD Lab (Reliable Adaptive Distributed Systems Lab). Dissertation work on the design and implementation of the X-Trace framework for tracing the execution of large,wide-area distributed systems [N1]. Deployed X-Trace in a number of systems, including the Coral Content Distribution Network running on Planetlab, where it was used to find bugs and diagnose performance problems. Also Rodrigo Fonseca 1 / 7 worked with professors Ion Stoica, Scott Shenker, and David Culler on the Sensor Network Architecture group, on novel networking protocols [S1, S3, S6] and architectures [S4, S5] for Wireless Sensor Networks. Jun 2004 − Hewlett-Packard Labs, Palo Alto- Research Intern Worked with Sujata Baner-
Cited by 0