Physics and Engineering of Radiation Detection

575

Chapter 10

Software for Data Analysis

With the advent of electronic detectors and fast data acquisition systems, the amount

of data available for analysis has grown tremendously. A natural consequence of

these developments has been the application of sophisticated analysis techniques

to the datasets. However this would not had been possible if there were not an

exponential increase in the availability of affordable computing power and memory.

Today’s computers perform millions of floating point operations in a flash, something

that was unthinkable before computers became available.

In earlier days of personal computing, the trend was to develop data analysis

packages from scratch and according to the particular needs. Most of these pack-

ages were developed in the programming languagefortran, which was and still is

a very powerful language. The power of fortran comes from its compiler that has

been specifically designed for scientific computing. Later on, as more and more

program routines were made available by the community to the general user, the

trend shifted towards implementing these small program snippets to develop the

required software. This trend is still continuing, though with a bit of twist now.

That is, now the developers use C++classesinstead of fortransubroutines.These

classes arereusableprograms, which can be modified by the programmer according

to the specific needs without re-coding. This powerful concept is the heart of the

object-orientedprogramming languages.

10.1StandardAnalysisPackages

In this section we will visit some of the commonly used data analysis package that

are available to the general user. Before we start, one thing that should be pointed

out is that the description of these software packages are not meant to be complete

in any sense and no attempt has been made to extensively test or compare the

packages. Each package has its own advantages and disadvantages and therefore

we do not recommend any one over the others. To make an informed decision,

the reader should go through the manuals of these packages as available on their

respective web sites and elsewhere.

10.1.AROOT...............................

ROOT is a very powerful data analysis and visualization package. It was originally

developed for the high energy physics community by Rene Brun and Fons Rademak-

ers at CERN in mid the 1990’s. An important part of root is its C++ interpretor