Parallel data acquisition for visualization of very large sparse matrices

An Efficient Parallel Algorithm for Extreme Eigenvalues of Sparse Nonsymmetric Matrices

The International Journal of Supercomputing Applications ◽

10.1177/109434209200600106 ◽

1992 ◽

Vol 6 (1) ◽

pp. 98-111 ◽

Cited By ~ 2

Author(s):

S. K. Kim ◽

A. T. Chrortopoulos

Keyword(s):

Shared Memory ◽

Message Passing ◽

Sparse Matrices ◽

Data Locality ◽

Main Memory ◽

Global Memory ◽

Global Communication ◽

Step Method ◽

Arnoldi Algorithm ◽

Large Sparse Matrices

Main memory accesses for shared-memory systems or global communications (synchronizations) in message passing systems decrease the computation speed. In this paper, the standard Arnoldi algorithm for approximating a small number of eigenvalues, with largest (or smallest) real parts for nonsymmetric large sparse matrices, is restructured so that only one synchronization point is required; that is, one global communication in a message passing distributed-memory machine or one global memory sweep in a shared-memory machine per each iteration is required. We also introduce an s-step Arnoldi method for finding a few eigenvalues of nonsymmetric large sparse matrices. This method generates reduction matrices that are similar to those generated by the standard method. One iteration of the s-step Arnoldi algorithm corresponds to s iterations of the standard Arnoldi algorithm. The s-step method has improved data locality, minimized global communication, and superior parallel properties. These algorithms are implemented on a 64-node NCUBE/7 Hypercube and a CRAY-2, and performance results are presented.

Download Full-text

Numerical test of stability of large sparse matrices and dynamical systems

Electronics Letters ◽

10.1049/el:19740400 ◽

1974 ◽

Vol 10 (23) ◽

pp. 503

Author(s):

V. Zakian ◽

C. Papaconstantinou

Keyword(s):

Dynamical Systems ◽

Sparse Matrices ◽

Numerical Test ◽

Large Sparse Matrices

Download Full-text

Proving the Preclusion of Data Manipulation Using Parallel Data Acquisition in Chromatography

Materials Science Forum ◽

10.4028/www.scientific.net/msf.941.2390 ◽

2018 ◽

Vol 941 ◽

pp. 2390-2394

Author(s):

David Thomas Marehn ◽

Detlef Wilhelm ◽

Heike Pospisil ◽

Roberto Pizzoferrato

Keyword(s):

Data Acquisition ◽

Best Practice ◽

Sampling Rate ◽

Good Manufacturing Practice ◽

Special Role ◽

Analytical Instruments ◽

Data Collector ◽

Parallel Data ◽

Time Variations ◽

Hplc Detector

Traceability has an enormous value for companies, but especially for those working in the regulated environment. It plays a special role in the field of pharmacy with respect to manufacturing, controlling and distributing batches of drugs. Through the guidance of Good Manufacturing Practice (GMP) traceability should be ensured. An increasing number of pharmaceutical companies are member of one of the global pharmacopoeias (United States Pharmacopeia, European Pharmacopeia and Japanese Pharmacopeia). The specifications of these pharmacopoeias describe the best practice in documentation, control, qualification and risk management. But however, the pharmacopoeias are written very generally and do not distinguish between the vendors of the analytical instruments. Here, we analyze how chromatographic analyses and data acquisition rely on a specific vendor of the device and the chromatography data system (CDS), the controlling software. We present a way to compare the data acquisition of different CDSs communicating with HPLC instruments. A newly developed software called Data Collector allows the acquisition of data from a HPLC detector parallel to the controlling CDS in the same run. Two HPLC systems and two different CDSs using a well defined sample standard have been tested. The direct comparison of the acquired data precludes unexpected data manipulations of both tested CDSs and shows that there are primarily deviations between the CDSs due to time variations only which depend on the sampling rate. All in all the Data Collector can be used for the traceability of data acquisition.

Download Full-text