By Topic

Workstation Operating Systems, 1992. Proceedings., Third Workshop on

Date 23-24 April 1992

Filter Results

Displaying Results 1 - 25 of 32
  • Proceedings. Third Workshop on Workstation Operating Systems (Cat. No.92TH0413-5)

    Publication Year: 1992
    Save to Project icon | Request Permissions | PDF file iconPDF (140 KB)  
    Freely Available from IEEE
  • Opal: a single address space system for 64-bit architecture address space

    Publication Year: 1992 , Page(s): 80 - 85
    Cited by:  Papers (11)  |  Patents (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (440 KB)  

    The recent appearance of architectures with flat 64-bit virtual addressing opens an opportunity to reconsider the way in which operating systems use virtual address spaces. An operating system called Opal is being built for these wide-address architectures. The key feature of Opal is a single global virtual address space that extends to data on long-term storage and across the network. The case for the use of a single virtual address space is outlined, the model of addressing and protection used in Opal is presented, and some of the problems and opportunities raised by the approach are discussed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The LITTLE WORK project

    Publication Year: 1992 , Page(s): 11 - 14
    Cited by:  Papers (9)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (276 KB)  

    Supporting software has not kept pace with the micro-miniaturization of microprocessor-based machines. The predominant operating system on such machines is MS-DOS, with no support for distributed computing. The goal of the LITTLE WORK project is to use off-the-shelf components to develop a mobile computing environment that is identical to that encountered in the office environment. Components identified as critical building blocks are the computer, the operating system, the file system, and communication. Each is examined. Development of a prototype machine is described View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The case for a new approach to operating systems for personal computers and work stations

    Publication Year: 1992 , Page(s): 164 - 167
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (208 KB)  

    The question of what precisely is meant by an operating system is briefly addressed, and a summary of the development of the operating station is given. Specific research challenges are outlined, and an RISC technique for operating system design is presented. Security questions are posed, and a series of discussion questions is provided View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Transparently interposing user code at the system interface

    Publication Year: 1992 , Page(s): 98 - 103
    Cited by:  Papers (1)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (340 KB)  

    The author reports on the development of an object-oriented toolkit which substantially increases the ease of interposing user code between clients and instances of the system interface by allowing such code to be written in terms of the high-level objects provided by this interface, rather than in terms of the intercepted system calls themselves. This toolkit helps enable new interposition agents to be written which otherwise would not have been attempted. It is used to construct several agents including protected environments for running untrusted binaries, modified file system namespaces, logical devices implemented entirely in user space, transparent network data compression and/or encryption agents, and system call tracing tools. Examples of other agents which could be built include transactional software environments and emulators for other operating system environments View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Issues in wireless mobile computing

    Publication Year: 1992 , Page(s): 2 - 10
    Cited by:  Papers (27)  |  Patents (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (552 KB)  

    The trend toward possible computers with interfaces to wireless network technologies is examined. The question of needed research and specification of hardware is discussed, the major challenges, such as services for the mobile user, mobile internetworking and adjusting to new technologies are described, and a listing of outstanding questions which must be resolved is presented View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • SPECmarks are leading us astray

    Publication Year: 1992 , Page(s): 160 - 161
    Cited by:  Papers (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (120 KB)  

    It is suggested that success at designing systems to satisfy latency-sensitive requirements has not yet been achieved, and that it is necessary to develop benchmarks that explicitly measure system latency, rather than measuring it indirectly through its effect on throughput. If both throughput-sensitive and latency-sensitive benchmarks are required, the result might be the creation of systems that are easily tuned (or even self-tuned) to provide either good latency or good throughput, as the case demands View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On kernel support for real-time multimedia applications

    Publication Year: 1992 , Page(s): 39 - 46
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (496 KB)  

    Real-time operating system services are required to support multimedia systems that rely heavily on the workstation processor for control of the audio and video processors and movement of audio and video data. The requirements for each service are described, together with the YARTOS kernel, an operating system kernel that provides real-time communication and computation services. The programming model supported by YARTOS is an extension of Wirth's discipline of real-time programming. In essence it is a message-passing system with a semantics of interprocess communication that specifies the real-time response that an operating system must provide to a message receiver. This allows a programmer to assert an upper bound on the time to receipt and processing of each message. The YARTOS kernel supports the notion of guaranteed processing rates. The desired processing rate of each task is made known to the kernel, and the kernel provides a guaranteed response time to each task that is sufficient for ensuring that the required processing rate is achieved View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Remote memory as a resource in distributed systems

    Publication Year: 1992 , Page(s): 132 - 136
    Cited by:  Papers (1)  |  Patents (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (328 KB)  

    To reduce reliance on disk I/O, the remote memory model is extended, such that dedicated server machines with substantial amounts of memory provide backing storage to client workstations. Remote memory provides the traditional service of swap space, but at faster access speeds. The remote memory server manages objects created, named, and destroyed by client machines. Objects are named by clients and can be shared by multiple clients. The memory server keeps track of idle memory available on other machines, shipping objects to those machines as needed. The memory server is operating-system- and architecture-independent, allowing all network clients to use the memory server. Thus, performance of the overall system can be improved by adding memory to the shared server View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The workstation as a waystation:-integrating mobility into computing environments

    Publication Year: 1992 , Page(s): 17 - 21
    Cited by:  Papers (2)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (236 KB)  

    With mobile computers, and increasingly mobile users, a new paradigm for integrated execution must be developed. An environment in which users may move freely among machines and may move machines freely from place to place, is described. In such an environment, a workstation is essentially a computer that temporarily provides windows and processing power to a user. The specific implications of mobility, i.e., window systems, process management, internet user naming, network software, and server reconfiguration are discussed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Device control abstractions for workstation operating system

    Publication Year: 1992 , Page(s): 34 - 38
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (284 KB)  

    Unpredictable workloads on a general-purpose workstation operating system will cause disturbances in process scheduling that become visible to the user. Systems that combine real-time and general-purpose characteristics in a single operating system have met with mixed success. The advent of multiprocessor workstations makes it feasible to dedicate functionally specific behavior to separate processing units. Going beyond the intelligent I/O controller, a user-programmable device interface would provide the control abstractions necessary for the efficient manipulation of high-speed devices. The connection to the workstation operating system would still mimic the conventional device interface of open/close/read/write/control, but would include a channel for downloading control programming that adapts a device for user specific applications View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Escaping the disk bottleneck in fast transaction processing

    Publication Year: 1992 , Page(s): 118 - 121
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (196 KB)  

    An approach to avoiding the disk access bottleneck for transaction processing applications is described. The approach is based on the use of primary copy replication, and takes advantage of recent hardware advances such as inexpensive high-speed CPUs and networks, and uninterruptable power supplies. In addition to improving response time, the architecture increases overall availability and reliability. The work described is in a preliminary stage. The overall architecture and its motivation are discussed. File system techniques in the context of transaction processing are reported View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Networking performance for microkernels

    Publication Year: 1992 , Page(s): 154 - 159
    Cited by:  Papers (4)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (368 KB)  

    Performance measurements for network protocols in microkernel systems are typically two to five times slower than comparable macrokernel systems. It is shown that user-level network protocols have performed poorly because they rely on code designed to run in a kernel environment. As a result, they make assumptions about the costs of primitive protocol operations such as scheduling, preemption, and data transfer which can require substantial overhead to satisfy at user level. Good user-level protocol performance can be achieved by restructuring protocol servers to take advantage of microkernel facilities, rather than ignore them View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Shared memory ought to be commonplace

    Publication Year: 1992 , Page(s): 86 - 90
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (392 KB)  

    Shared memory as a programming abstraction is widely used within parallel applications. It is not widely used between applications. It is suggested that shared memory is both faster and more intuitive than the principal alternatives in many cases, and that the general disuse of shared memory facilities in systems such as Unix is due in large part to a lack of appropriate tools. A series of measures to make shared memory more convenient is being pursued. Dynamic linking is used to allow programs to access shared or persistent data in the same way they access ordinary variables and functions. Memory and files are unified into a single-level store that facilitates the sharing of pointers and capitalizes on emerging 64-bit architectures. Existing interfaces and tools are exploited to remain backward-compatible with Unix View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Toward massive distributed file systems

    Publication Year: 1992 , Page(s): 48 - 51
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (308 KB)  

    Massive scale in distributed file systems remains an elusive goal. Existing systems are generally limited to only a few hundred clients or make restrictive assumptions, such as that widely shared files are read-only. It is argued that a truly massive system must scale for all kinds of files; file access traces suggest that occasionally written files make up too large a proportion of shared reads to be ignored. The file access patterns that are likely to be important in the large-scale systems of the future are outlined, and a scheme for constructing dynamic client hierarchies that may work on a much greater scale than conventional approaches is described View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Lost in a labyrinth of workstations

    Publication Year: 1992 , Page(s): 52 - 55
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (248 KB)  

    Two efforts at descriptive name service for workstations are discussed. The Networked Resource Discovery Project proposes to use signposts to locate resources. This approach is effective in the search for one of many instances of a service. Nomenclator is a system that combines maps with signs. Early Nomenclator results indicate that there are significant performance advantages to be had by using a combination of signs and maps to locate resources. If this approach to descriptive name services is pursued, it is argued that users will not become lost in a labyrinth of workstations View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A plea for interfaces that support caching

    Publication Year: 1992 , Page(s): 137 - 140
    Cited by:  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (208 KB)  

    The kinds of problems client designers face when using interfaces without adequate caching support are discussed. A description of more recent file server protocols in which caching support is present is given. The difficulties interface designers will face in incorporating caching support are described. The designer faces the problem of making apparent the basic functionality of the interface, uncluttered by caching support, and the problem of what kind of caching support to provide. It is argued that incorporating caching support in the interfaces found in modern operating systems and distributed systems is critical to the performance and functionality of workstation applications View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Workstation autonomy is a dead issue

    Publication Year: 1992 , Page(s): 15 - 16
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (128 KB)  

    Two significant questions relating to workstation autonomy are control of resources and control of information. It is argued that in light of developing technology these factors will become increasingly irrelevant over the next decade, and that work will be possible from anonymous workstations with the computation load split appropriately between one or more processors local to the workstation and additional processors available via the network. It will be essential that writing programs for this environment be as straightforward as possible. Scheduling of work across such a distributed system will involve both global decisions (process placement and migration) and local ones. In computer servers with heavily cached memory systems, the appropriate granularity for local scheduling will be coarser than for interactive workstations. This raises issues not only for the local scheduler, but also for the application partitioning tools View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Workstation cooperation through a typed distributed shared memory abstraction

    Publication Year: 1992 , Page(s): 70 - 74
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (324 KB)  

    Data units are defined in a level that is above simple memory but below other cooperation paradigms. They are easily ported and well matched to traditional programming. Implementations can vary to suit each system, but the interface remains the same. They can be added to various systems without affecting the operating system context. Data units make an excellent foundation for building fully functional objects. However, they do not render the classic procedural programming model obsolete. Various coherency policies and mechanisms, including those of classic distributed shared memory (DSM) can be applied to data units. Data unit locks permit a wide variety of application-controlled coherency. This is particularly useful for those applications which allow the use of stale data while a replica is being updated. With proper control, this can lead to improved concurrency View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The case for application-specific operating systems

    Publication Year: 1992 , Page(s): 92 - 94
    Cited by:  Papers (10)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (116 KB)  

    Recent evidence suggests that trends will require rethinking the traditional role of operating systems. The challenge to operating systems designers is to deliver to applications the performance available now only from dedicated hardware, combined with the ease of sharing resources and data among multiple applications and the simpler programming model found in general-purpose operating systems. An application-specific structure is proposed where as much of the operating system as possible is pushed into runtime library routines linked with each application. The operating system kernel is stripped to its bare minimum functionality. At a minimum, the kernel must adjudicate among application requests for physical resources, and it must enforce hardware protection boundaries by operating system code running as library routines in each application. The key is that the operating system must notify each application of changes in its resource allocation, to allow the application on the chance to adapt to make best use of whatever resources are available to it View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Supporting the information mesh

    Publication Year: 1992 , Page(s): 56 - 61
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (348 KB)  

    A model for the role of the network in a distributed computing environment that will necessarily require abstractions in the operating system has been developed. This model is based on information that is long lived and widely distributed. It assumes that interesting information can be distributed around the world, and survives outside any particular application, application toolkit, or programming language runtime system. Within the information mesh, computations or activities will take place in one of three forms: one in which the focus of attention is changed, but the application remains the same; one in which the focus changes, and the application itself must also change; and a third that is less specific, more global, e.g., searching. In the cases of such global operations, one may need to limit the scope of terms of the mesh, since these are operations over regions of the mesh. It is important to include links and a native form of link shadow in each operating system, if the workstation running that operating system is to allow the user to collaborate in the information mesh View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On the ubiquity of logging in distributed file systems

    Publication Year: 1992 , Page(s): 122 - 125
    Cited by:  Papers (1)  |  Patents (8)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (232 KB)  

    It is argued that logging should be at the forefront of techniques considered by a system designer when implementing a distributed file system. The use of logging in different guises in the Coda file system is described. Coda is a distributed file system whose goal is to provide highly available, scalable, secure and efficient shared file access in an environment of Unix workstations. High availability is achieved through two complementary mechanisms, i.e., server replication and disconnected operation. Logging is used in at least three distinct ways in the current implementation of Coda. First, value logging forms the basis of the recovery technique for recoverable virtual memory (RVM), a transactional virtual memory package. Second, operating logging is used in the replay log that records update activity made by a client while disconnected from all servers. Third, operation logging is used in resolution logs on severs to allow transparent resolution of directory updates made to partitioned server replicas View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Medium term virtual memory replacement

    Publication Year: 1992 , Page(s): 142 - 147
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (264 KB)  

    A measurement technique that is used to monitor memory usage is described, and two new algorithms for memory management are presented. The first is a virtual memory replacement algorithm that is partially based on periodic, sequential, and transient behaviors. The other is an approximation to the line replaceable unit (LRU), called the cluster LRU, that performs better on the programs measured than the usual clocl algorithm. The performance of various algorithms is compared by trace-drive simulation. The new algorithms seem to provide significant improvement over existing approaches. Faults decrease approximately 10-40% in the range of interest for the memory traces View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The benefits of service rebalancing

    Publication Year: 1992 , Page(s): 104 - 110
    Cited by:  Papers (1)  |  Patents (5)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (332 KB)  

    Service rebalancing, which provides a way to determine an efficient division of effort between a client and its server, is introduced. Decisions concerning this division of labor are made at runtime rather than at design time. Evaluating the current environment in which the client and server are executing and moving mode between client and server based on this evaluation can enhance the performance of client/server programs. The advantages of service rebalancing include the elimination of a static division between client and server, on-the-fly updating of modules, load balancing, sharing of common code between multiple clients, and the enforcement of neatly modularized programming. Some of the problems and issues related to service rebalancing, including equanimity and the current status of the work, are discussed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Disk reads with DRAM latency

    Publication Year: 1992 , Page(s): 126 - 131
    Cited by:  Papers (3)  |  Patents (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (356 KB)  

    To overcome the relative deterioration of secondary storage performance, input/output systems are being designed around highly parallel disk arrays, high-bandwidth optical networks, and cache-specialized file systems. Disk arrays increase secondary storage throughput by spreading data over many disks so that large accesses are striped for parallel operation and so that many small accesses can operate concurrently. Fast networks preserve throughput improvements derived from disk arrays on file server as data is delivered to increasingly distant client systems. Secondary storage latency is primarily addressed by large client and server file caches. Whenever data to be read is found in the cache or data to be written can be delayed in the cache, latency is as short as if secondary storage were constructed with semiconductor memory. The current research focuses on reducing the access latency for data to be read that is not already cached View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.