Putting It All Together：Experience and Challenges at the DELPHI off—line Processing Farm

After over 10 years of existence,DELPHI off-line software counts altogether over 1500k lines.Being written by multitude of authorsk,many of them having already left,it is very incoherent and extremely difficult to maintain:it is written in Fortran,it relies on obsolete tools and it has to run in a distributed multi-os computing environment.Still.as the analysis of LEP data will continue during the next 5-6 years,this code will have to be used and be ported to yet another platforms.In order to ensure high efficiency in use of our resources,we have developed several tools which hide from the user most of the intricaices of the operating system,batch system and data access.These tools are well integrated and easy to maintain.As the problems are quite typical for the High Energy Physics software,we believe that the ideas we have implemented can be useful also for the next generation of experiments.     

Development of the ATLAS Simulation Framework

Object-oriented (OO) approach is the key technology to develop a software system in the LHC/ATLAS experiment.We developed a OO simulation framework based on the Geant4 general-purpose simulation toolkit.Because of complexity of simulation in ATLAS,we payed most attention to the scalability in its design.Although the first target to apply this framework is to implement the ATLAS full detector simulation program,there is no experiment-specific code in it,therefore it can be utilized for the development of any simulation package,not only for HEP experiments but also for various different research domains ,In this paper we discuss our approach of design and implementation of the framework.     

基于ActiveX机制的VB与MATLAB数据交换

MATLAB是目前应用最广泛的数值科学计算语言．同时也是系统分析和仿真的强大工具．VB是目前使用最广泛的可视化编程工具．本文介绍一种利用ActiveX机制实现VB与MATLAB数据交换的方法．从而可以充分结合两个软件的优点．大大缩短软件开发周期．     

基于PC-Linux的STAR TOFp探测器的数据获取系统

廉价的运行开放源代码操作系统Linux的硬件在高能物理领域中正扮演着各种各样重要的角色.本文描述了为STARTOFp探测器设计的基于PC-Linux的数据获取系统.系统采用了常规的解决方案:由一台运行Linux的PC来控制由NIM和CAMAC插件构成的前端电子学处理部分.系统已经被成功的集成进STAR数据获取系统,并且在STAR的第二年的运行中工作正常.     

集成管理软件及其开发策略

随着科技的进步,尤其以信息技术、通讯技术为核心的高新技术的蓬勃兴起,使现代企业的管理面临着新的挑战。 一、计算机企业管理的最新发展     

PW1700X射线衍射仪织构分析应用软件的功能开发

由Ｐｈｉｌｉｐｓ引进的ＰＷ１７００Ｘ射线衍射仪织构测量系统只能用反射法，按螺旋线扫测极目数据，测绘极角。α≤８５°的不完整极图。经功能开发后，建立了织构分析测算系统，织构分析应用软件包，包括不完整极图测量与ＯＤＦ（取向分布函数）计算，回算全极图与反极图，可彩色屏显考贝，并能进行织构的定量分析。适用于立方和六方品系多晶织构材料分析。近两年应用表明，软件功能优异。     

The CDF Run 2 Offline Computer Farms

Run 2 at Fermilab began in March,2001,CDF will collect data at a maximum rate of 20 MByte/sec during the run.The offline reconstruction of this data must keep up with the data taking rate.This reconstruction occurs on a large PC farm,which must have the capacity for quasi-real time data reconstruction,for reprocessing of some data and for generation and processing of Monte Carlo samples.In this paer we will give the design requirements ofr the farm,describe the hardware and software design used to meet those requirements,describe the early experiences with Run 2 data processing,and discussfuture prospects for the farm,including some ideas about Run 2b processing.     

Abstract Interfaces for Data Analysis —Component Architecture for Data Analysis Tools

The fast turnover of software technologies,in particular in the domain of in teractivity(covering user interface and visualisation)makes it difficult for a small group of people to produce complete and polished software-tools before the underlying technologies make them obsolete.At the HepVis ‘99 workshop,a working group has been formed to improve the rpoduction of software tools for data analysis in HENP.Beside promoting a distributed development organisation,one goal of the group is to systematically design a set of abstract interfaces based on using modern OO analysis and OO design techniques.An initial domain analysis has come up with several categories(componets)found in typical data analysis tools:historams,Ntuples,Functions,Vectors,Fitter,Plotter,Analyzer and Controller,Special Emphasis was put on reducing the couplings between the categories to a minimum,thus optimising re-use and maintainability of any component individually.The interfaces have been defined in Java and C and implementations exist in the form of libraries and tools using C (Anaphe/Lizard,Openscientist)and Java(Java Analysis Studio),A special implementation aims at accessing the Java Liraries(through their Abstract Interfaces)from C .This paper giver an overview of the architecture and design of the various components for data analysis as discussed in AIDA.     

The Linux Based Distributed Data Acquisition System for the ISTRA＋ Experiment

The DAQ hardware of the ISTRA experiment consists of the VME system crate that contains two PCI-VME bridges interfacing two PC‘s with VME,external interrupts receiver,the readout controller for dedicated front-end electronics,the read-out controller buffer memory module,the VME-CAMAC interface,and additional control modules,The DAQ computing consist of 6 PC‘s running the Linux operating system and linked into LAN.The first PC serves the external interrupts and acquires the data from front-end electronics,The second one is the slow control computer.The remaining PC‘s host the monitoring and data analysis software.The Linux based DAQ software provides the external interrupts procssing,the data acquisition,recording,and distribution between monitoring and data analysis tasks running at DAQ PC‘s.The monitoring programs are based on two packages for data visualization:home-written on and the ROOT system.My SQl is used as a DAQ database.