Unix-like Operating Systems Research Articles

The Quiet Rebellion

This article discusses Linux which is an emerging technology that will affect many mechanical engineers. Linux is a freely distributed, open-source Unix-like operating system, originally for PCs, but now available on other platforms. The Internet facilitated the collaborative effort that produced Linux—and so it is not surprising that the explosive growth of the Internet was also made possible by Linux. Linux would be a fantastic operating system for building automation systems and direct digital controls. One area where Linux really shines is Local Area Networking, using a program called Samba (which is included in almost all Linux distributions). The technical support for Linux is very good. Chances are very good that no matter where you are, there is a local Linux users group nearby where you can turn for help. These groups often have mailing lists that you can subscribe to, and post queries. Linux resources on the Internet are outstanding. Local users groups will occasionally have “install fests,” where you can get help installing Linux on your PC.

Multi-threading: A new dimension to massively parallel scientific computation

Multi-threading is becoming widely available for Unix-like operating systems, and the application of multi-threading opens new ways for performing parallel computations with greater efficiency. We here briefly discuss the principles of multi-threading and illustrate the application of multi-threading for a massively parallel direct four-index transformation of electron repulsion integrals. Finally, other potential applications of multi-threading in scientific computing are outlined.

HERA-B higher-level triggers: architecture and software

HERA-B will be studying CP-violation in the B-system in a high-rate hadronic environment. To accomplish this goal, HERA-B needs a sophisticated data acquisition and trigger system. Except for the first level, all trigger levels are implemented as PC-farms, running the Unix-like operating system, Linux, thus blurring the sharp border between online and offline application software. The hardware architecture and software environments are discussed.

Hands-on software design [real-time programming

Reactive and time-critical software is used when safety is the issue and the margin for errors is narrow. The issues of timing, concurrency, interprocess communication, resource sharing and external device handling are very important. Graduates who understand how the software will interact with the operating system and the environment are in high demand. In an effort to provide students with such knowledge, the Embry-Riddle Aeronautical University has been offering real-time courses. These courses use the standard UNIX platform and Ada as the implementation language to support concurrency and software engineering practices. A dedicated real-time laboratory was also designed. The laboratory contains hardware (complete computers and interface devices) for teaching both soft and hard real-time concepts. Resources include a wide variety of hardware platforms operating different UNIX-like operating systems. The real-time laboratory supports time-critical software development with full pre-emptiveness and deterministic behavior. Using the laboratory, the students can experiment. This aids in learning real-time concepts. The experiments address such concepts as: reactive and time-critical programming, concurrency of programming tasks, multiprocessing and multithreaded applications, signals and operating system interfaces, resource contention constructs, the implementation of concurrent programs, communication protocols and reactive input/output interfaces.

Nonintrusive Cloning Garbage Collection with Stock Operating System Support

It is well accepted that automatic garbage collection simplifies programming, promotes modularity, and reduces development effort. However it is commonly believed that these advantages do not counteract the perceived price: excessive overheads, possible long pause times while garbage collections occur, and the need to modify existing code. Even though there are publically available garbage collector implementations that can be used in existing programs, they do not guarantee short pauses, and some modification of the application using them is still required. In this paper we describe a snapshot-at-beginning concurrent garbage collector algorithm and its implementation. This algorithm guarantees short pauses, and can be easily implemented on stock UNIX-like operating systems. Our results show that our collector performs comparable to other garbage collection implementations on uniprocessor machines and outperforms similar collectors on multiprocessor machines. We also show our collector to be competitive in performance with explicit deallocation. Our collector has the added advantage of being non-intrusive. Using a dynamic linking technique and effective root set inferencing, we have been able to successfully run our collector even in commercial programs where only the binary executable and no source code is available. In this paper we describe our algorithm, its implementation, and provide both an algorithmic and a performance comparison between our collector and other similar garbage collectors. ©1997 by John Wiley & Sons, Ltd.

A migration tool to support resource and load sharing in heterogeneous computing environments

Distributed computing systems potentially provide significant advantages and benefits such as enhanced performance, extensibility, reliability, and better resource and load sharing. However, the evolution of hardware and software has unavoidably led to heterogeneous computing systems. The heterogeneity of such systems gives rise to many problems such as transparency, application developments, database and network management, efficient resource and load sharing, migration, etc. This paper discusses the problems associated with resource and load sharing in such heterogeneous environments. First, it discusses different techniques such as distributed file systems, network operating systems, load-sharing algorithms, migration mechanisms and other methods which have been proposed as solutions for resource and load sharing. Then it describes our design and implementation of a prototype tool for program and data migration in heterogeneous computing environments of different computer architectures and different operating systems. The tool was based on the client–server model and the use of remote procedure calls for interconnecting heterogeneous distributed computer systems. This important decision supports the evolution of the computer hardware and software without side-effects. It adopts well with the evolution of distributed systems as the degree of heterogeneity increases. The tool aims to allow better resource and load sharing in heterogeneous computer systems; to allow useful information to be quickly obtained; to reduce communication costs and programmers' time in prototyping distributed applications. It is designed and implemented on a local network of computers connected by Ethernet and running different Unix and Unix-like operating systems.

Filters: QoS support mechanisms for multipeer communications

The nature of distributed multimedia applications is such that they require multipeer communication support mechanisms. The multimedia traffic needs to be delivered to end-systems, networks, and end-users in a form that they can handle while satisfying the constraints imposed by the multimedia application. Quality-of-service (QoS) mechanisms that can ensure full quality media playout at high-performance workstations, while at the same time providing appropriately filtered lower quality media for playout at other end-systems, are required. Existing multicast support mechanisms are deficient for this purpose, in a heterogeneous environment, because they work on a common denominator premise where the quality delivered depends on the least capable link or node involved in the multicast session. This paper begins by discussing video compression; it proposes and analyzes the use of filtering mechanisms as means of supporting disparate receiver capabilities and QoS requirements. The paper describes the implementation of a number of filtering mechanisms and highlights the communications architecture within which these mechanisms are built. This architecture constitutes a specific network topology and a new protocol family developed within a UNIX-like operating system.

The equipment access software for a distributed UNIX-based accelerator control system

This paper presents a generic equipment access software package for a distributed control system using computers with UNIX or UNIX-like operating systems. The package consists of three main components, an application Equipment Access Library, Message Handler and Equipment Data Base. An application task, which may run in any computer in the network, sends requests to access equipment through Equipment Library calls. The basic request is in the form Equipment-Action-Data and is routed via a remote procedure call to the computer to which the given equipment is connected. In this computer the request is received by the Message Handler. According to the type of the equipment connection, the Message Handler either passes the request to the specific process software in the same computer or forwards it to a lower level network of equipment controllers using MIL1553B, GPIB, RS232 or BITBUS communication. The answer is then returned to the calling application. Descriptive information required for request routing and processing is stored in the real-time Equipment Data Base. The package has been written to be portable and is currently available on DEC Ultrix, LynxOS, HPUX, XENIX, OS-9 and Apollo domain.

Linkage in the Nemesis single address space operating system

The recent interest in single address space operating systems has resulted in a number of papers, most of which gloss over the issues of linking programs to run in multiple protection domains. Some of the confusion about 64-bit address spaces is due to the almost pervasive use of UNIX and UNIX-like operating systems (such as Mach, Chorus and Amoeba) and languages with poor enforcement of abstraction like C and C++.This paper describes some of the linkage structure of Nemesis, a multi-service operating system being developed as part of the Pegasus project. Nemesis provides a simple and efficient mechanism for program linkage which provides rich sharing of text at a level of individual object classes.

Signaling and operating system support for native-mode ATM applications

Applications communicating over connectionless networks, such as IP, cannot obtain per-connection Quality of Service (QoS) guarantees. In contrast, the connection-oriented nature of the ATM layer and its per-virtual-circuit QoS guarantees are visible to a native-mode ATM application. We describe the design and implementation of operating system and signaling support for native-mode applications, independent of the semantics of the protocol layers or of the signaling protocol. The work was done in the context of a Unix-like operating system and the Xunet 2 wide-area high-speed ATM network. The IPC-based interface between an application and the signaling entity allows processes to request parameterized virtual circuits, and the signaling-kernel interface allows resources to be reclaimed from prematurely terminating processes. We also built a simple encapsulation layer over raw IP that allows any host with IP access to send AAL frames into the wide-area network with little performance degradation. Our design makes it simple to port existing TCP/IP socket applications to a native-mode ATM protocol stack and also enables interoperation of existing IP networks with our ATM network. Our experience has been positive - the design is robust, easily extendible and scales well with the number of open connections.

Linda meets Minix

Currently large interest is devoted to the development of systems supporting parallel and distributed programming over a network of workstations. We are exploring operating system support for this kind of programming, that instead is usually implemented at the user process level. We have chosen the Linda model as the basis of a programming system for networks of personal computers running Minix, a Unix-like operating system for personal computers.In this paper we shortly describe the Minix operating system and its services for remote communication based on the Amoeba protocol. We have extended Minix with new communication primitives based on Linda; we show their implementation, describing the new system architecture, and compare the new system with other Linda implementations.

Natural language interface based on keyword extraction using AWK

The paper describes a method of implementing a natural language interface based on keyword extraction using the AWK programming language. This method is suitable for designing natural language interfaces on systems running under Unix or Unix-like operating systems, for applications such as computer-aided instruction and database retrieval. The knowledge base and software implementation of the natural language interface are described.

An object-oriented operating system interface

This paper discusses an object-oriented interface from the Smalltalk-80™ programming environment to a Unix-like operating system. This interface imposes an object-oriented paradigm on operating system facilities. We discuss some of the higher order abstractions that were created to make use of these facilities, and discuss difficulties we encountered implementing this interface. Several examples of cooperating Smalltalk and operating system processes are presented.