SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Cooperative processing for the 1990's using client-server technology is addressed. The main theme is concepts of downsizing from mainframes and minicomputers to workstations on a local area network (LAN). This document is presented in view graph form.
A brief introduction to the Universal Naming Protocol (UNP) is presented. UNP specifies a generic attribute-based name server upon which a variety of high-level naming services, including white- and yellow-page services, can be built.
File systems have long been the most important and most widely used form of shared permanent storage. File systems in traditional time-sharing systems, such as Unix, support a coherent sharing model for multiple users. Distributed file systems implement this sharing model in local area networks. However, most distributed file systems fail to scale from local area networks to an internet. Four characteristics of scalability were recognized: size, wide area, autonomy, and heterogeneity. Owing to size and wide area, techniques such as broadcasting, central control, and central resources, which are widely adopted by local area network file systems, are not adequate for an internet file system. An internet file system must also support the notion of autonomy because an internet is made up by a collection of independent organizations. Finally, heterogeneity is the nature of an internet file system, not only because of its size, but also because of the autonomy of the organizations in an internet. The Jade File System, which provides a uniform way to name and access files in the internet environment, is presented. Jade is a logical system that integrates a heterogeneous collection of existing file systems, where heterogeneous means that the underlying file systems support different file access protocols. Because of autonomy, Jade is designed under the restriction that the underlying file systems may not be modified. In order to avoid the complexity of maintaining an internet-wide, global name space, Jade permits each user to define a private name space. In Jade's design, we pay careful attention to avoiding unnecessary network messages between clients and file servers in order to achieve acceptable performance. Jade's name space supports two novel features: (1) it allows multiple file systems to be mounted under one direction; and (2) it permits one logical name space to mount other logical name spaces. A prototype of Jade was implemented to examine and validate its design. The prototype consists of interfaces to the Unix File System, the Sun Network File System, and the File Transfer Protocol.
The Sprite operating system allows executing processes to be moved between hosts at any time. We use this process migration mechanism to offload work onto idle machines, and also to evict migrated processes when idle workstations are reclaimed by their owners. Sprite's migration mechanism provides a high degree of transparency both for migrated processes and for users. Idle machines are identified, and eviction is invoked, automatically by daemon processes. On Sprite it takes up to a few hundred milliseconds on SPARCstation 1 workstations to perform a remote exec, while evictions typically occur in a few seconds. The pmake program uses remote invocation to invoke tasks concurrently. Compilations commonly obtain speedup factors in the range of three to six; they are limited primarily by contention for centralized resources such as file servers. CPU-bound tasks such as simulations can make more effective use of idle hosts, obtaining as much as eight-fold speedup over a period of hours. Process migration has been in regular service for over two years.
JPL is under contract, through NASA, with the US Army to develop a state-of-the-art Command Center System for the US European Command (USEUCOM). The Command Center System will receive, process, and integrate force status information from various sources and provide this integrated information to staff officers and decision makers in a format designed to enhance user comprehension and utility. The system is based on distributed workstation class microcomputers, VAX- and SUN-based data servers, and interfaces to existing military mainframe systems and communication networks. JPL is developing the Command Center System utilizing an incremental delivery methodology called the Rapid Development Methodology with adherence to government and industry standards including the UNIX operating system, X Windows, OSF/Motif, and the Ada programming language. Through a combination of software engineering techniques specific to the Ada programming language and the Rapid Development Approach, JPL was able to deliver capability to the military user incrementally, with comparable quality and improved economies of projects developed under more traditional software intensive system implementation methodologies.
Rapid progress in low-cost commercial PC-class multimedia workstation technology will potentially have a dramatic impact on the productivity of distributed work groups of 50-100 software developers. Hypermedia/multimedia involves the seamless integration in a graphical user interface (GUI) of a wide variety of data structures, including high-resolution graphics, maps, images, voice, and full-motion video. Hypermedia will normally require the manipulation of large dynamic files for which relational data base technology and SQL servers are essential. Basic machine architecture, special-purpose video boards, video equipment, optical memory, software needed for animation, network technology, and the anticipated increase in productivity that will result for the introduction of hypermedia technology are covered. It is suggested that the cost of the hardware and software to support an individual multimedia workstation will be on the order of $10,000.
A process-oriented control model is described for distributed problem solving. The model coordinates the transfer and manipulation of information across independent networked applications, both intelligent and conventional. The model was implemented using SOCIAL, a set of object-oriented tools for distributing computing. Complex sequences of distributed tasks are specified in terms of high level scripts. Scripts are executed by SOCIAL objects called Manager Agents, which realize an intelligent coordination model that routes individual tasks to suitable server applications across the network. These tools are illustrated in a prototype distributed system for decision support of ground operations for NASA's Space Shuttle fleet.
Five scientists at MSFC/ESAD have EOS SCF investigator status. Each SCF has unique tasks which require the establishment of a computing facility dedicated to accomplishing those tasks. A SCF Working Group was established at ESAD with the charter of defining the computing requirements of the individual SCFs and recommending options for meeting these requirements. The primary goal of the working group was to determine which computing needs can be satisfied using either shared resources or separate but compatible resources, and which needs require unique individual resources. The requirements investigated included CPU-intensive vector and scalar processing, visualization, data storage, connectivity, and I/O peripherals. A review of computer industry directions and a market survey of computing hardware provided information regarding important industry standards and candidate computing platforms. It was determined that the total SCF computing requirements might be most effectively met using a hierarchy consisting of shared and individual resources. This hierarchy is composed of five major system types: (1) a supercomputer class vector processor; (2) a high-end scalar multiprocessor workstation; (3) a file server; (4) a few medium- to high-end visualization workstations; and (5) several low- to medium-range personal graphics workstations. Specific recommendations for meeting the needs of each of these types are presented.
Multicast services needed for current distributed applications on LAN's fall generally into one of three categories: datagram, semi-reliable, and reliable. Transport layer multicast datagrams represent unreliable service in which the transmitting context 'fires and forgets'. XTP executes these semantics when the MULTI and NOERR mode bits are both set. Distributing sensor data and other applications in which application-level error recovery strategies are appropriate benefit from the efficiency in multidestination delivery offered by datagram service. Semi-reliable service refers to multicasting in which the control algorithms of the transport layer--error, flow, and rate control--are used in transferring the multicast distribution to the set of receiving contexts, the multicast group. The multicast defined in XTP provides semi-reliable service. Since, under a semi-reliable service, joining a multicast group means listening on the group address and entails no coordination with other members, a semi-reliable facility can be used for communication between a client and a server group as well as true peer-to-peer group communication. Resource location in a LAN is an important application domain. The term 'semi-reliable' refers to the fact that group membership changes go undetected. No attempt is made to assess the current membership of the group at any time--before, during, or after--the data transfer.
The problem of routing jobs to parallel queues with identical exponential servers and unequal finite buffer capacities is considered. Stochastic ordering and weak majorization properties on critical performance measures are established by means of event-driven inductions. In particular, it is shown that the intuitive 'join the shortest non-full queue' (SNQ) policy is optimal with respect to an overall function that accounts for holding and blocking costs. Moreover, the buffer allocation problem is solved by proving the intuitive result that, for a fixed total buffer capacity, the optimal allocation scheme is the one in which the difference between the maximum and minimum queue capacities is minimized, i.e., becomes either 0 or 1.
The computing requirements of the individual scientific computing facilities (SCFs) have been evaluated. The requirements under consideration include CPU-intensive vector and scalar processing, visualization, data storage, connectivity, and I/O peripherals. A hierarchy consisting of shared and individual resources is proposed to most effectively meet the total SCF computing requirements. This hierarchy includes a supercomputer class vector processor; a high-end scalar multiprocessor workstation; a high-speed, large capacity file server; a few medium to high-end visualization workstations; and low to medium range personal graphics workstations.
There are now over one million UNIX sites and the pace at which new installations are added is steadily increasing. Along with this increase, comes a need to develop simple efficient, effective and adaptable ways of simultaneously collecting real-time diagnostic and performance data. This need exists because distributed systems can give rise to complex failure situations that are often un-identifiable with single-machine diagnostic software. The simultaneous collection of error and performance data is also important for research in failure prediction and error/performance studies. This paper introduces a portable method to concurrently collect real-time diagnostic and performance data on a distributed UNIX system. The combined diagnostic/performance data collection is implemented on a distributed multi-computer system using SUN4's as servers. The approach uses existing UNIX system facilities to gather system dependability information such as error and crash reports. In addition, performance data such as CPU utilization, disk usage, I/O transfer rate and network contention is also collected. In the future, the collected data will be used to identify dependability bottlenecks and to analyze the impact of failures on system performance.
The STX mass storage system design is intended for environments requiring high speed access to large volumes of data (terabyte and greater). Prior to commitment to a product design plan, STX conducted an exhaustive study of the commercially available off-the-shelf hardware and software. STX also conducted research into the area of emerging technologies in networks and storage media so that the design could easily accommodate new interfaces and peripherals as they came on the market. All the selected system elements were brought together in a demo suite sponsored jointly by STX and ALLIANT where the system elements were evaluated based on actual operation using a client-server mirror image configuration. Testing was conducted to assess the various component overheads and results were compared against vendor data claims. The resultant system, while adequate to meet our capacity requirements, fell short of transfer speed expectations. A product team lead by STX was assembled and chartered with solving the bottleneck issues. Optimization efforts yielded a 60 percent improvement in throughput performance. The ALLIANT computer platform provided the I/O flexibility needed to accommodate a multitude of peripheral interfaces including the following: up to twelve 25MB/s VME I/O channels; up to five HiPPI I/O full duplex channels; IPI-s, SCSI, SMD, and RAID disk array support; standard networking software support for TCP/IP, NFS, and FTP; open architecture based on standard RISC processors; and V.4/POSIX-based operating system (Concentrix). All components including the software are modular in design and can be reconfigured as needs and system uses change. Users can begin with a small system and add modules as needed in the field. Most add-ons can be accomplished seamlessly without revision, recompilation or re-linking of software.
The building blocks for an open storage system includes a system platform, a selection of storage devices and interfaces, system software, and storage applications CONVEX storage systems are based on the DS Series Data Server systems. These systems are a variant of the C3200 supercomputer with expanded I/O capabilities. These systems support a variety of medium and high speed interfaces to networks and peripherals. System software is provided in the form of ConvexOS, a POSIX compliant derivative of 4.3BSD UNIX. Storage applications include products such as UNITREE and EMASS. With the DS Series of storage systems, Convex has developed a set of products which provide open system solutions for storage management applications. The systems are highly modular, assembled from off the shelf components with industry standard interfaces. The C Series system architecture provides a stable base, with the performance and reliability of a general purpose platform. This combination of a proven system architecture with a variety of choices in peripherals and application software allows wide flexibility in configurations, and delivers the benefits of open systems to the mass storage world.
The simulation of the network control center (NCC) is in the second phase of development. This phase seeks to further develop the work performed in phase one. Phase one concentrated on the computer systems and interconnecting network. The focus of phase two will be the implementation of the network message dialogues and the resources controlled by the NCC. These resources are requested, initiated, monitored and analyzed via network messages. In the NCC network messages are presented in the form of packets that are routed across the network. These packets are generated, encoded, decoded and processed by the network host processors that generate and service the message traffic on the network that connects these hosts. As a result, the message traffic is used to characterize the work done by the NCC and the connected network. Phase one of the model development represented the NCC as a network of bi-directional single server queues and message generating sources. The generators represented the external segment processors. The served based queues represented the host processors. The NCC model consists of the internal and external processors which generate message traffic on the network that links these hosts. To fully realize the objective of phase two it is necessary to identify and model the processes in each internal processor. These processes live in the operating system of the internal host computers and handle tasks such as high speed message exchanging, ISN and NFE interface, event monitoring, network monitoring, and message logging. Inter process communication is achieved through the operating system facilities. The overall performance of the host is determined by its ability to service messages generated by both internal and external processors.
Access to online information sources of aerospace, scientific, and engineering data, a mission focus for NASA's Scientific and Technical Information Program, has always been limited to factors such as telecommunications, query language syntax, lack of standardization in the information, and the lack of adequate tools to assist in searching. Today, the NASA STI Program's NASA Access Mechanism (NAM) prototype offers a solution to these problems by providing the user with a set of tools that provide a graphical interface to remote, heterogeneous, and distributed information in a manner adaptable to both casual and expert users. Additionally, the NAM provides access to many Internet-based services such as Electronic Mail, the Wide Area Information Servers system, Peer Locating tools, and electronic bulletin boards.
Motivated by issues of weapon safety and security, laser based firing systems for promptly initiating secondary explosives have been under active development at Sandia National Laboratories for more than four years. Such a firing system consists of miniaturized, Q-switched, solid-state laser, optical detonators, optical safety switches, and elements for splitting, coupling, and transmitting the laser output. Potential system applications pose significant challenges in terms of server mechanical and thermal environments and packaging constraints, while requiring clear demonstration of safety enhancements. The Direct Optical Initiation (DOI) Program at Sandia is addressing these challenges through progress development phases during which the design, fabrication, and testing of prototype hardware is aimed at more difficult application requirements. A brief history of the development program, and a summary of current and planned activities, will be presented.
We describe the implementation of a large scale disk array controller and subsystem incorporating over 100 high performance 3.5 inch disk drives. It is designed to provide 40 MB/s sustained performance and 40 GB capacity in three 19 inch racks. The array controller forms an integral part of a file server that attaches to a Gb/s local area network. The controller implements a high bandwidth interconnect between an interleaved memory, an XOR calculation engine, the network interface (HIPPI), and the disk interfaces (SCSI). The system is now functionally operational, and we are tuning its performance. We review the design decisions, history, and lessons learned from this three year university implementation effort to construct a truly large scale system assembly.