Red Hat Linux Research Articles

Computer simulation and calibration of the charge particle spectrometer-telescope ⪡STEP-F⪢

The spectrometer-telescope “STEP-F” is intended for detecting high-energy charged particle fluxes in near-Earth space. It has been developed to study solar cosmic rays and dynamics of the Earth's radiation belts onboard the Russian spacecraft ⪡Coronas-Photon⪢. The results of simulations by the CERN GEANT4.2 Code under OS LINUX Red Hat 6.2 for some types of particles, which will be detected by the “STEP-F” instrument, are presented. The visualization of primary and secondary particle tracks in each detector of the telescope has allowed us to verify the energy ranges of the recorded particles and to ascertain the possibility to determine the direction of incidence of primary electrons. The detectors have been tested in a high-energy ion cyclotron accelerator at the Institute of Physical and Chemical Research (RIKEN, Japan) in order to determine the response of each of the detectors to primary α-particle and hydrogen ion H2 beams. The experimental data obtained are in good agreement with the results of Monte-Carlo simulations. The basic parameters of the “STEP-F” instrument, such as the energy range of recorded electrons, protons and α-particles, the geometric factor, the field of view and the angular resolution, are presented.

Secure UNIX socket-based controlling system for high-throughput protein crystallography experiments.

A control system for high-throughput protein crystallography experiments has been developed based on a multilevel secure (SSL v2/v3) UNIX socket under the Linux operating system. Main features of protein crystallography experiments (purification, crystallization, loop preparation, data collecting, data processing) are dealt with by the software. All information necessary to perform protein crystallography experiments is stored (except raw X-ray data, that are stored in Network File Server) in a relational database (MySQL). The system consists of several servers and clients. TCP/IP secure UNIX sockets with four predefined behaviors [(a) listening to a request followed by a reply, (b) sending a request and waiting for a reply, (c) listening to a broadcast message, and (d) sending a broadcast message] support communications between all servers and clients allowing one to control experiments, view data, edit experimental conditions and perform data processing remotely. The usage of the interface software is well suited for developing well organized control software with a hierarchical structure of different software units (Gaponov et al., 1998), which will pass and receive different types of information. All communication is divided into two parts: low and top levels. Large and complicated control tasks are split into several smaller ones, which can be processed by control clients independently. For communicating with experimental equipment (beamline optical elements, robots, and specialized experimental equipment etc.), the STARS server, developed at the Photon Factory, is used (Kosuge et al., 2002). The STARS server allows any application with an open socket to be connected with any other clients that control experimental equipment. Majority of the source code is written in C/C++. GUI modules of the system were built mainly using Glade user interface builder for GTK+ and Gnome under Red Hat Linux 7.1 operating system.

ASEP/MD: A program for the calculation of solvent effects combining QM/MM methods and the mean field approximation

ASEP/MD is a computer program designed to implement the Averaged Solvent Electrostatic Potential/Molecular Dynamics (ASEP/MD) method developed by our group. It can be used for the study of solvent effects and properties of molecules in their liquid state or in solution. It is written in the FORTRAN90 programming language, and should be easy to follow, understand, maintain and modify. Given the nature of the ASEP/MD method, external programs are needed for the quantum calculations and molecular dynamics simulations. The present version of ASEP/MD includes interface routines for the GAUSSIAN package, HONDO, and MOLDY, but adding support for other programs is straightforward. This article describes the program and its usage. Program summary Title of program: ASEP/MD Catalogue identifier:ADSF Program Summary URL: http://cpc.cs.qub.ac.uk/summaries/ADSF Program obtainable from: CPC Program Library, Queen's University of Belfast, N. Ireland Computer for which the program is designed: it has been tested on Intel-based PC and Sun Operating systems under which the program has been tested: Red Hat Linux 7.2 and SunOS 5.6 Programming language used: FORTRAN90 Memory required to execute with typical data: greatly depends on the system No. of processors used: 1 Has the code been vectorized or parallelized?: no No. of bytes in distributed program, including test data, etc.: 44 544 Distribution format: tar gzip file Keywords: Solvent effects, QM/MM methods, mean field approximation, geometry optimization Nature of physical problem: The study of molecules in solution with quantum methods is a difficult task because of the large number of molecules and configurations that must be taken into account. The quantum mechanics/molecular mechanics methods proposed to date either require massive computational power or oversimplify the solute quantum description. Method of solution: A non-traditional QM/MM method based on the mean field approximation was developed where a classical molecular dynamics simulation is coupled with a quantum calculation. The average electrostatic potential generated by the solvent over the solute is calculated from the simulation and introduced into the quantum calculation as an external field. This process can be performed iteratively. Standard external programs are used for the molecular dynamics simulations and for the quantum calculations. The present program acts as an interface and controls the flow of the calculation. Restrictions on the complexity of the problem: At present, only pure liquids and binary dilute solutions (a single solute molecule) can be studied. For the molecular dynamics only MOLDY is implemented, while GAUSSIAN and HONDO are available for the quantum calculations. Restrictions of the aforementioned programs apply. Typical running time: Running time depends on the nature of the chemical system and the options passed to the external programs, which are usually by far the longest part of the calculations. Unusual features of the program: Uses SYSTEM and GETARG calls.

The Rutgers Computational Grid: A Distributed Linux PC Cluster

The Rutgers Computational Grid (RCG) project is aimed at providing high throughput performance to Rutgers university faculty and students. The RCG employs dual processor PCs, with Pentium II and III processors, as computational nodes, running the Linux RedHat operating system. The Load Sharing Facility (LSF) scheduling system from Platform Computing is used for job control and monitoring. The nodes are grouped into subclusters physically located in several departments and controlled by a single master node through LSF. The hardware and software used in RCG are described. Utilization and performance issues, including parallel performance, are discussed based on the experience of the first two years of RCG operation.

SIGCSE 2002 forum

The intent of this forum is to generate a really lively discussion on the gamut of our relationship - as computer science educators - to industry.How dependent are we/should we be on specific industry software and platforms? (Many examples come to mind - Microsoft Visual Studio.NET, Red Hat Linux, C#, Sun's version of Java.)How obligated are we to jump on board the newest upgraded software version of whatever we might be using?How does our desire for good job placement of our graduates square with our oft-repeated claim that we are educators, not trainers?Pursuing this issue of "where do we fall on the spectrum between training and education," what to do about all these industry certification programs - Cisco, Novell, Oracle, Microsoft. Do we teach these curricula? Do we give credit to students who take it, say at a community college, and then transfer?What about a service course in, say, Photoshop? Is this college-level material?Do opportunities for summer internships or research projects in an industry setting place some special obligations on us with respect to that company?

Building a cyberwar lab: lessons learned

With funding from NSF the Department has set up a stand alone lab for students to learn penetration testing techniques(attack), to harden their networks against these attacks (defense) , and also to gather enough evidence to through careful logging and audit controls to convict a hacker (convict). Linux RedHat 7.1 was used and all the machines were set up as standalone servers in three different subdomains, with 2 perimeter routers and 2 firewalls to allow experimentation with various configurations. In all over 50 software tools were downloaded and tested. Students were screened and asked to sign a disclaimer. They should also have been required to have networking experience. An initial mistake was to run a very minimal server with no services and practically no users. This was not realistic. It made it quick to rebuild systems but much harder to attack.The attacks need to be carefully planned and structured in a specific sequence one at a time, otherwise it becomes very difficult to follow what is going on.

The performance of a supercomputer built with commodity components

We built a supercomputer called Galaxy by connecting Intel Pentium-based computer nodes with Fast and Gigabit Ethernet switches. Each node has two processors at clock speeds varying from 300 to 600 MHz, up to 512 MB of memory, and small 2 Gb local disk. All nodes run the standard RedHat Linux and inter-node communication is handled by a message passing interface called MPI. Local tools are written to visualize the system performance and to balance loads. We have benchmarked a sub-Galaxy with 72 processors by NAS and Parallel LINPACK benchmark suites. We achieved 16.9 Gflops in a standard single precision LU decomposition for 46848×46848 matrix parallel LINPACK benchmark. A Galaxy with 128 processors costs approximately $250 000 and it delivers 40 Gflops of performance. This leads to a cost-performance ratio of 160 Kflops-per-dollar, which is to improve further due to increase in processor speeds and network bandwidth at similar cost. Our final system with 512 processors is expected to reach several Tflops. This article first describes the Galaxy architectural details, and then present and analyze its performance in terms of floating point number crunching, network bandwidth, and IO throughput.

The future of the OS for internet applications

While the main focus of Internetrelated OS activity is on Unix and Windows, a growing number of organizations are using the open-source Linux OS in their servers. Red Hat recently released Red Hat Linux 6.1, which is designed for better performance, scalability, and installation than earlier versions. Thus, no company controls the market for the OS for Internet applications.