Physics Data Analysis Research Articles

NASA Space Physics Program

The selection of proposals to be funded under NASA's Space Physics Division Research and Analysis (R&A) Program in fiscal year (FY) 1989 has been completed, and the results of this selection are summarized here.These proposals were submitted in response to NASA Research Announcement (NRA) NRA‐88‐OSSA‐02, issued in February of this year. The NRA covered most but not all of the Division's R&A programs, as can be seen in Table 1. Basic R&A programs in Heliopsheric Physics, Ionospheric Physics, Magnetospheric Physics, and Solar Physics were reviewed, along with the Cosmic Physics Theory and Data Analysis. Not included in this review were the experimental portion of the Cosmic Physics program, the Suborbital Program (sounding rockets and balloons), and the Solar‐Terrestrial Theory Program.

The use of color and gray-scale scatter-plots in the analysis of high-energy physics data

We present results of tests of the standard model of electroweak interactions from the AMY, TOPAZ and VENUS detectors at TRISTAN operating at centre of mass energies up to 60.8 GeV. Measured asymmetries in e+e− → μ+μ−, τ+μ− and bb are consistent with the standard model. The hadronic event rate is higher than expected for Mz = 92 GeV/c2, but is consistent with the standard model with Mz = 88.5 GeV/c2, sin2Θw = 0.234, and ΛMS = 0.15 GeV.

Data driven parallelism in experimental high energy physics applications

I present global design principles for the implementation of High Energy Physics data analysis code on sequential and parallel processors with mixed shared and local memory. Potential parallelism in the structure of High Energy Physics tasks is identified with granularity varying from a few times 10 8 instructions all the way down to a few times 10 4 instructions. It follows the hierarchical structure of detector and data acquisition systems. To take advantage of this - yet preserving the necessary portability of the code - I propose a computational model with purely data driven concurrency in Single Program Multiple Data (SPMD) mode. The Task granularity is defined by varying the granularity of the central data structure manipulated. Concurrent processes coordinate themselves asynchroneously using simple lock constructs on parts of the data structure. Load balancing among processes occurs naturally. The scheme allows to map the internal layout of the data structure closely onto the layout of local and shared memory in a parallel architecture. It thus allows to optimize the application with respect to synchronization as well as data transport overheads. I present a coarse top level design for a portable implementation of this scheme on sequential machines, multiprocessor mainframes (e.g. IBM 3090), tightly coupled multiprocessors (e.g. RP-3) and loosely coupled processor arrays (e.g. LCAP, Emulating Processor Farms).

The scope of APL in nuclear measurements

The whole range of data acquisition and analysis in experimental nuclear physics can be covered with a few APL application workspaces. Typical steps of the treatment of statistical data are highlighted and compared with nearly-equivalent solutions in other languages.

A Multiple Processor System for Acquisition and Analysis of Nuclear Physics Data

A data acquisition/analysis system for nuclear physics experiments using an array of three single-board computers has been constructed and tested. The computers, each containing a Motorola 68000 CPU, are coupled together on the commercially available VMEbus, and communicate to a host PDP-11/55 via a DMA link. A specialized operating system has been written for the multiprocessor array, which works in conjunction with an existing generalpurpose data acquisition program. The system has been in use for several months at the University of Pennsylvania Tandem Accelerator Laboratory for on-line experiments and replaying of data. In actual operation, each processor provides approximately the same computing power as a PDP-11/55, and the execution speed of the array is found to be proportional to the number of processors used.

Use of the Fermilab Advanced Computer Program Multi-Microprocessor as an on-Line Trigger Processor

The Fermilab Advanced Computer Program is constructing a powerful multi-microprocessor system for data analysis in high-energy physics. The system will use commercial 32-bit microprocessors and be programmed in Fortran-77. A large array of support software allows easy migration of user applications from a uniprocessor environment to the multiprocessor, as well as providing sophisticated program development, debugging, and error handling and recavery tools. This system will provide computing power with a cost effectiveness of $3000-4000 per VAX 11/780 equivalent. We describe here the applications of this system as an on-line trigger processor. The low cost, commercial availability, compatibility with off-line analysis programs, and high data bandwidths (up to 160 MByte/sec) make this system an ideal choice for such real-time applications.

Dual Display System for Multidimensional Data Analysis of High-Energy Physics

We have constructed an interactive graphic system with two cathode ray tube displays which we have called the Dual Display System (DDS) for quick and efficient data analysis in high energy physics. We used the new system to analyze data from 6 GeV/c π-p inelastic reactions and found that the system could separate the reaction channels more cleanly than conventional methods and thus provide reliable information on the reaction mechanisms. This paper describes the new system and presents the software techniques for multidimensional analysis used with it.

KEKNET, a high speed on-line network for high energy physics

A high speed on-line network for the data acquisition and analysis in the high energy physics developed at KEK is described. Its design principle and operational results are presented.

Application of an Array Processor to Data Analysis in Medium Energy Physics

The rate at which data may be acquired using LAMPF's high resolution magnetic spectrometers, HRS and EPICS, at times considerably exceeds the rate at which it may be analyzed with the on-line PDP-11/45's. At HRS in particular, data is routinely taken in a mode where less than 20% of the events may be completely analyzed even using well optimzed software. In order to achieve a situation in which data may be analyzed at a rate comparable to that at which it was acquired, a Floating Point Systems AP-120B Array Processor has been interfaced to the off-line PDP-11/60. The goal of this combination is to produce a system on which data may be analyzed at least a factor of 5 faster than is possible on-line, while maintaining as nearly as possible the same capabilities and user interface as with the standard off-line or on-line analysis.

RT/RSX MULTI: Packages for Data Acquisition and Analysis in High-Energy Physics

The MULTI system is a software package for data acquisition, data analysis, and experiment control. PDP-11 implementations are in widespread use under the RT-11 and RSX-1lM operating systems at Fermilab and elsewhere. The system philosophy, architecture, and significant features are discussed, and the measured performance is reported.

New Methods for Derivation of Plasma Light Data from Television Apparatus

We describe some novel uses of television for continuous monitoring and adjustment of a pulsed beam plasma experiment. Quantitative information can be obtained giving the position, size, and shape of the light emitting plasma. The absolute light intensity and the profile of the plasma can be separately obtained from a video scan analyzer. Time resolution of events occurring in a time span of ∼10 nsec can be achieved by incorporating a pulsed image intensifier ahead of the television camera tube. These techniques supplant photographic methods and augment information derivable from microwave and magnetic diagnostics. Some suggestions are made for future applications to other types of data analysis in physics and astronomy.

The Bell-Rutgers Data Highway System

The Bell-Rutgers data highway system is a modular, general purpose system designed to provide a framework in which the configuration and reconfiguration of on-line computer hardware is easily accomplished. It is built using modified NIM hardware, and features a 16 bit parallel data path for input and output which is shared by up to 150 general purpose modules. Special consideration has been given to two particular problems; the operation of modules in an environment in which digital switching noise must be minimized, and the simultaneous use of the module system by several independent users. The initial application of the system has been in the implementation of a computer based nuclear physics data acquisition and analysis system. The following paper is intended as an exposition of some of the considerations which are relevant to the design of any general purpose modular system as well as a description of a particular system. Some of the ways in which the system has been used are also described.

Authors and Papers

Authors and Papers

A computer facility for nuclear physics data collection and analysis

A medium size digital computer is used in a time-shared mode for real-time data collection and independent background computation.