Scientific Linux
Abstract
In the midst of the well advertised battle for the server and desktop
markets, Linux has quietly become a dominant platform in the scientific and
engineering markets.
With a powerful combination of open source code, low costs, excellent
performance, minimal porting requirements from other Unix variants and support
for a wide range of platforms, Linux has emerged as a significant computer
solution for the scientific community.
This presentation examines the key strengths of Linux in science and
engineering, focusing on:
- The reasons why Linux has become a key platform for scientific work,
- The benefits of access to kernel source code for scientific software
developers,
- The broad range of free scientific software now available for
Linux, including development tools, numerical libraries, visualisation
tools and scientific programs such as SciLab, Rasmol and Octave,
- The use of Linux as a cost-effective scientific workstation in a
heterogeneous computing and networking environment, often replacing
high-end workstations (running other Unix variants or Windows NT) as
large "number crunching" programs are ported across to Linux
- The Beowulf project: high-performance computing at the right price,
allowing researchers with minimal funds to create supercomputing clusters
from commodity computer parts,
- The Linux Lab Project, which acts as a clearing house for Linux kernel
drivers and interfacing software used to communicate with data acquisition
and control hardware such as A/D cards, timer/counter cards and digitisers,
- Real-time Linux: a layer of kernel extensions that provide real-time
capabilities for precise data acquisition and hardware control, without
requiring extensive modifications to either the Linux kernel or user
software,
- LabView for Linux: National Instruments have released a Linux version of
the de facto industry standard for instrument interfacing and virtual
instrumentation,
- Embedded Linux: using Linux for telemetry, systems control and SCADA
interfacing in a wide range of scientific, engineering and
telecommunications applications, on a variety of computer platforms such
as ARM-based wearable computers and embedded PC104 boards, and
- A summary of other significant scientific Linux resources, such as the
SAL (Scientific Applications on Linux) website
These points will be illustrated with a number of examples from around the
world, including our own work at Monash University developing Linux-based
wastewater monitors for Sydney Water Corporation.
Next
Updated: 20 June 1999