Complexity Management and Design Optimization Regarding a Variety of Triple Modular Redundancy Schemes through Automation
No abstract available
SEARCH · Search NASA
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.
No abstract available
Commercial digital signal processors (DSPs) for use in high-speed satellite computers are challenged by the damaging effects of space radiation, mainly single event upsets (SEUs) and single event functional interrupts (SEFIs). Innovations have been developed for mitigating the effects of SEUs and SEFIs, enabling the use of very-highspeed commercial DSPs with improved SEU tolerances. Time-triple modular redundancy (TTMR) is a method of applying traditional triple modular redundancy on a single processor, exploiting the VLIW (very long instruction word) class of parallel processors. TTMR improves SEU rates substantially. SEFIs are solved by a SEFI-hardened core circuit, external to the microprocessor. It monitors the health of the processor, and if a SEFI occurs, forces the processor to return to performance through a series of escalating events. TTMR and hardened-core solutions were developed for both DSPs and reconfigurable field-programmable gate arrays (FPGAs). This includes advancement of TTMR algorithms for DSPs and reconfigurable FPGAs, plus a rad-hard, hardened-core integrated circuit that services both the DSP and FPGA. Additionally, a combined DSP and FPGA board architecture was fully developed into a rad-hard engineering product. This technology enables use of commercial off-the-shelf (COTS) DSPs in computers for satellite and other space applications, allowing rapid deployment at a much lower cost. Traditional rad-hard space computers are very expensive and typically have long lead times. These computers are either based on traditional rad-hard processors, which have extremely low computational performance, or triple modular redundant (TMR) FPGA arrays, which suffer from power and complexity issues. Even more frustrating is that the TMR arrays of FPGAs require a fixed, external rad-hard voting element, thereby causing them to lose much of their reconfiguration capability and in some cases significant speed reduction. The benefits of COTS high-performance signal processing include significant increase in onboard science data processing, enabling orders of magnitude reduction in required communication bandwidth for science data return, orders of magnitude improvement in onboard mission planning and critical decision making, and the ability to rapidly respond to changing mission environments, thus enabling opportunistic science and orders of magnitude reduction in the cost of mission operations through reduction of required staff. Additional benefits of COTS-based, high-performance signal processing include the ability to leverage considerable commercial and academic investments in advanced computing tools, techniques, and infra structure, and the familiarity of the science and IT community with these computing environments.
A simple heuristic algorithm for designing highly reliable modularly redundant computer systems under complexity constraints is presented. The technique, which produces near optimal solutions, is intuitively appealing and easy to apply. The algorithms performance is shown to compare very well with the optimal solution obtained via a computerized model for dynamic programming.
Electric utilities have a main responsibility to protect the lives and safety of their workers when they are working on low-, medium-, and high-voltage power lines and distribution circuits. With the anticipated widespread deployment of smart grids, a secure and highly reliable means of maintaining isolation of customer-owned distributed generation (DG) from the affected distribution circuits during maintenance is necessary to provide a fully de-energized work area, ensure utility personnel safety, and prevent hazards that can lead to accidents such as accidental electrocution from unanticipated power sources. Some circuits are serviced while energized (live line work) while others are de-energized for maintenance. For servicing de-energized circuits and equipment, lock-out tag-out (LOTO) programs provide a verifiable procedure for ensuring that circuit breakers are locked in the off state and tagged to indicate that status to operational personnel so that the lines will be checked for voltage to verify they are de-energized. The de-energized area is isolated from any energized sources, which traditionally are the substations. This procedure works well when all power sources and their interconnections are known armed with this knowledge, utility personnel can determine the appropriate circuits to de-energize for isolating the target line or equipment. However, with customer-owned DG tied into the grid, the risk of inadvertently reenergizing a circuit increases because circuit connections may not be adequately documented and are not under the direct control of the local utility. Thus, the active device may not be properly de-energized or isolated from the work area. Further, a remote means of de-energizing and locking out energized devices provides an opportunity for greatly reduced safety risk to utility personnel compared to manual operations. In this paper, we present a remotely controllable LOTO system that allows individual workers to determine the configuration and status of electrical system circuits and permit them to lock out customer-owned DG devices for safety purposes using a highly secure and ultra-reliable radio signal. The system consists of: (1) individual personal lockout devices, (2) lockout communications and logic module at circuit breakers, which are located at all DG devices, and (3) a database and configuration control process located at the utility operations center. The lockout system is a close permissive, i.e., loss of control power or communications will cause the circuit breaker to open. Once the DG device is tripped open, a visual means will provide confirmation of a loss of voltage and current that verifies the disconnected status of the DG. Further the utility personnel will be able to place their own lock electronically on the system to ensure a lockout functionally. The proposed LOTO system provides enhanced worker safety and protection against unintended energized lines when DG is present. The main approaches and challenges encountered through designing the proposed region-wide LOTO system are discussed in this paper. These approaches include: (1) evaluating the reliability of the proposed approach under N-modular redundancy with voter/spares configurations and (2) conducting a system level risk assessment study using the failure modes and effects analysis (FMEA) technique to identify and rank failure modes by probability of occurrence, probability of detection, and severity of consequences. This ranking allows a cost benefits analysis to be conducted such that dollars and efforts will be applied to the failures that provide greatest incremental gains in system capability (resilience, survivability, security, reliability, availability, etc.) per dollar spent whether capital, operations, or investment. Several simulation scenarios and their results are presented to demonstrate the viability of these approaches.
System combines two concepts - triple modular redundancy (TMR) and spares (replacement system) redundancy. System has triplicated active unit core with associated bank of specifically arranged standby spare units. Curves plotted from characteristic equation show hybrid system is more reliable than conventional TMR systems.
This paper presents a modular-redundant Message Passing Interface (MPI) solution, MR-MPI, for transparently executing high-performance computing (HPC) applications in a redundant fashion. The presented work addresses the deficiencies of recovery-oriented HPC, i.e., checkpoint/restart to/from a parallel file system, at extreme scale by adding the redundancy approach to the HPC resilience portfolio. It utilizes the MPI performance tool interface, PMPI, to transparently intercept MPI calls from an application and to hide all redundancy-related mechanisms. A redundantly executed application runs with r*m native MPI processes, where r is the number of MPI ranks visible to the application and m is the replication degree. Messages between redundant nodes are replicated. Partial replication for tunable resilience is supported. The performance results clearly show the negative impact of the O(m^m) messages between replicas. For low-level, point-to-point benchmarks, the impact can be as high as the replication degree. For applications, performance highly depends on the actual communication types and counts. On single-core systems, the overhead can be 0% for embarrassingly parallel applications independent of the employed redundancy configuration or up to 70-90% for communication-intensive applications in a dual-redundant configuration. On multi-core systems, the overhead can be significantly higher due to the additional communication contention.
For fault-tolerant computer systems consisting of multiple classes of modules, a unified method for analyzing mission reliability is proposed and evaluated. The analysis proceeds by generalizing the notions of standby and N modular redundancy into a concept called hybrid-degraded redundancy. The probabilistic evaluation of the unified redundancy concept is then developed to yield, for a given modular class, the joint distribution of success and the number of nonfailed modules from that class, at special times. With this information, a Markov chain analysis gives the reliability of an entire sequence of phases (mission profile).
Design options for decreasing the susceptibility of integrated circuits to Single Event Upset (SEU) fall into two categories: (1) increasing the critical charge to cause an upset at a particular node, and (2) employing redundancy to mask or correct errors. With decreasing device sizes on an Integrated Circuit (IC), the amount of charge required to represent a logic state has steadily reduced. Critical charge methods such as increasing drive strength or increasing the time required to change state as in capacitive or resistive hardening or delay based approaches extract a steadily increasing penalty as a percentage of device resources and performance. Dual redundancy is commonly assumed only to provide error detection with Triple Modular Redundancy (TMR) required for correction, but less well known methods employ dual redundancy to achieve full error correction by voting two inputs with a prior state to resolve ambiguity. This requires special circuits such as the Whitaker latch [1], or the guard-gate [2] which some of us have called a Transition AND Gate (TAG) [3]. A 2-input guard gate is shown in Figure 1. It is similar to a Muller Completion Element [4] and relies on capacitance at node "out" to retain the prior state when inputs disagree, while eliminating any output buffer which would be susceptible to radiation strikes. This paper experimentally compares delay based and dual rail flip-flop designs wherein both types of circuits employ guard-gates to optimize layout and performance, and draws conclusions about design criteria and suitability of each option. In both cases a design goal is protection against Single Event Transients (SET) in combinational logic as well as SEU in the storage elements. For the delay based design, it is also a goal to allow asynchronous clear or preset inputs on the storage elements, which are often not available in radiation tolerant designs.
Since 1974, CNES has been involved in geostationary positioning. Among different entities participating in operations and their preparation, the Flight Dynamics Center (FDC) is in charge of performing the following tasks: orbit determination; attitude determination; computation, monitoring, and calibration of orbit maneuvers; computation, monitoring, and calibration of attitude maneuvers; and operational predictions. In order to fulfill this mission, the FDC receives telemetry from the satellite and localization measurements from ground stations (e.g., CNES, NASA, INTELSAT). These data are processed by space dynamics programs integrated in the MERCATOR system which is run on SUN workstations (UNIX O.S.). The main features of MERCATOR are redundancy, modularity, and flexibility: efficient, flexible, and user friendly man-machine interface; and four identical SUN stations redundantly linked in an Ethernet network. Each workstation can perform all the tasks from data acquisition to computation results dissemination through a video network. A team of four engineers can handle the space mechanics aspects of a complete geostationary positioning from the injection into a transfer orbit to the final maneuvers in the station-keeping window. MERCATOR has been or is to be used for operations related to more than ten geostationary positionings. Initially developed for geostationary satellites, MERCATOR's methodology was also used for satellite control centers and can be applied to a wide range of satellites and to future manned missions.
Telerobotic rovers equipped with adequate actuators and sensors are clearly necessary for extraterrestrial construction. They will be employed as substitutes for humans, to perform jobs like surveying, sensing, signaling, manipulating, and the handling of small materials. Important design criteria for these rovers include versatility and robustness. They must be easily programmed and reprogrammed to perform a wide variety of different functions, and they must be robust so that construction work will not be jeopardized by parts failures. The key qualities and functions necessary for these rovers to achieve the required versatility and robustness are modularity, redundancy, and coordination. Three robotic rovers are being built by CSC as a test bed to implement the concepts of modularity and coordination. The specific goal of the design and construction of these robots is to demonstrate the software modularity and multirobot control algorithms required for the physical manipulation of constructible elements. Each rover consists of a transporter platform, bus manager, simple manipulator, and positioning receivers. These robots will be controlled from a central control console via a radio-frequency local area network (LAN). To date, one prototype transporter platform frame was built with batteries, motors, a prototype single-motor controller, and two prototype internal LAN boards. Software modules were developed in C language for monitor functions, i/o, and parallel port usage in each computer board. Also completed are the fabrication of half of the required number of computer boards, the procurement of 19.2 Kbaud RF modems for inter-robot communications, and the simulation of processing requirements for positioning receivers. In addition to the robotic platform, the fabrication of a local positioning system based on infrared signals is nearly completed. This positioning system will make the rovers into a moving reference system capable of performing site surveys. In addition, a four degree mechanical manipulator especially suited for coordinated teleoperation was conceptually designed and is currently being analyzed. This manipulator will be integrated into the rovers as their end effector. Twenty internal LAN cards fabricated by a commercial firm are being used, a prototype manipulator and a range finder for a positioning system were built, a prototype two-motor controller was designed, and one of the robots is performing its first telerobotic motion. In addition, the robots' internal LAN's were coordinated and tested, hardware design upgrades based on fabrication and fit experience were completed, and the positioning system is running. The rover system is able to perform simple tasks such as sensing and signaling; coordination systems which allow construction tasks to begin were established, and soon coordinated teams of robots in the laboratory will be able to manipulate common objects.
A soft error-mitigating semiconductor design system and associated methods that tailor circuit design steps to mitigate corruption of data in storage elements (e.g., flip flops) due to Single Events Effects (SEEs). Required storage elements are automatically mapped to triplicated redundant nodes controlled by a voting element that enforces majority-voting logic for fault-free output (i.e., Triple Modular Redundancy (TMR)). Storage elements are also optimally positioned for placement in keeping with SEE-tolerant spacing constraints. Additionally, clock delay insertion (employing either a single global clock or clock triplication) in the TMR specification may introduce useful skew that protects against glitch propagation through the designed device. The resultant layout generated from the TMR configuration may relax constraints imposed on register transfer level (RTL) engineers to make rad-hard designs, as automation introduces TMR storage registers, memory element spacing, and clock delay/triplication with minimal designer input.
A soft error-mitigating semiconductor design system and associated methods that tailor circuit design steps to mitigate corruption of data in storage elements (e.g., flip flops) due to Single Events Effects (SEEs). Required storage elements are automatically mapped to triplicated redundant nodes controlled by a voting element that enforces majority-voting logic for fault-free output (i.e., Triple Modular Redundancy (TMR)). Storage elements are also optimally positioned for placement in keeping with SEE-tolerant spacing constraints. Additionally, clock delay insertion (employing either a single global clock or clock triplication) in the TMR specification may introduce useful skew that protects against glitch propagation through the designed device. The resultant layout generated from the TMR configuration may relax constraints imposed on register transfer level (RTL) engineers to make rad-hard designs, as automation introduces TMR storage registers, memory element spacing, and clock delay/triplication with minimal designer input.
Neuromorphic processors are designed to execute Deep Neural Networks (DNNs) at very high speed using only a fraction of the electrical power needed to run a DNN on a traditional CPU or GPU. This unique capability makes Neuromorphic processors a prime candidate for space systems, where advanced computational tasks like image analysis, depth map reconstruction, or rover control need to be executed in a power-starved environment. In contrast to the growing number of applications of Neuromorphic processors in smart phones, the automotive and robotics domain, the space environment is unforgiving because of extreme temperatures and high levels of radiation. Any space system, operating beyond LEO requires computing hardware that is resilient against radiation effects. However, Neuromorphic processors have not yet been designed or tested for their radiation tolerance. In this report, we consider traditional methods of detection of radiation events and mitigation via redundancy and gauge their effectiveness on DNNs. In contrast to traditional flight software, however, neural networks represent a statistical algorithm, which might affect its resilience against radiation events. We will focus on the analysis of the tolerance of DNNs with respect to radiation events and discuss techniques to detect radiation hits using on-chip triple modular redundancy (TMR) on an Intel Loihi neuromorphic processor and to mitigate radiation damage. We describe an architecture for on-chip TMR for the Intel Loihi and present results of initial experiments.
Traditional avionics computing systems are federated, with each system provided on a number of dedicated hardware units. Federated applications are physically separated from one another and analysis of the systems is undertaken individually. Integrated Modular Avionics (IMA) takes these federated functions and integrates them on a common computing platform in a tightly deterministic distributed real-time network of computing modules in which the different applications can run. IMA supports different levels of criticality in the same computing resource and provides a platform for implementation of fault tolerance through hardware and application redundancy. Modular implementation has distinct benefits in design, testing and system maintainability. This paper covers the requirements for fault tolerant bus systems used to provide reliable communication between IMA computing modules. An overview of the Time Triggered Protocol (TTP) specification and implementation as a reliable solution for IMA systems is presented. Application examples in aircraft avionics and a development system for future space application are covered. The commercially available TTP controller can be also be implemented in an FPGA and the results from implementation studies are covered. Finally future direction for the application of TTP and related development activities are presented.
In this paper we report the development of a technique for modeling transient faults in redundant computer systems. Transient faults are characterized by their arrival rate and their duration. Fault detection, transient recovery, and the effect of permanent faults are included. A fault occurrence/recovery status state diagram is drawn to illustrate the operational status of the system while undergoing faults. The state diagram is used to formulate the equations for the mission failure probability. The techniques are then applied to a triple modular redundant computer system.
Design of the nuclear thermionic space power system, 40 50 70 Kw(e) power range, are given. The design configuration (1) meets the constraints of readily available launch vehicles; (2) allows for off-design operation including startup, shutdown, and possible emergency conditions; (3) provides tolerance of failure by extensive use of modular, redundant elements; (4) incorporates and uses heat pipes in a fashion that reduces the need for extensive in-pile testing of system components; and (5) uses thermionic converters, nuclear fuel elements, and heat transfer devices in a geometrical form adapted from existing incore thermionic system designs. Designs and in some cases performance data for elements and groups of the elements of the system are included. Benefits of the highly modular system approach to reliability, safety, economy of development, and flexibility are discussed.
The concept of designing for reliability will be introduced along with a brief overview of reliability, redundancy and traditional methods of fault tolerance is presented, as applied to current logic devices. The fundamentals of advanced circuit design and analysis techniques will be the primary focus. The introduction will cover the definitions of key device parameters and how analysis is used to prove circuit correctness. Basic design techniques such as synchronous vs asynchronous design, metastable state resolution time/arbiter design, and finite state machine structure/implementation will be reviewed. Advanced topics will be explored such as skew-tolerant circuit design, the use of triple-modular redundancy and circuit hazards, device transients and preventative circuit design, lock-up states in finite state machines generated by logic synthesizers, device transient characteristics, radiation mitigation techniques. worst-case analysis, the use of timing analyzer and simulators, and others. Case studies and lessons learned from spaceflight designs will be given as examples
The geometric rate of improvement of transistor size and integrated circuit performance known as Moore's Law has been an engine of growth for our economy, enabling new products and services, creating new value and wealth, increasing safety, and removing menial tasks from our daily lives. Affordable, highly integrated components have enabled both life-saving technologies and rich entertainment applications. Anti-lock brakes, insulin monitors, and GPS-enabled emergency response systems save lives. Cell phones, internet appliances, virtual worlds, realistic video games, and mp3 players enrich our lives and connect us together. Over the past 40 years of silicon scaling, the increasing capabilities of inexpensive computation have transformed our society through automation and ubiquitous communications. Looking forward, increasing unpredictability threatens our ability to continue scaling integrated circuits at Moore's Law rates. As the transistors and wires that make up integrated circuits become smaller, they display both greater differences in behavior among devices designed to be identical and greater vulnerability to transient and permanent faults. Conventional design techniques expend energy to tolerate this unpredictability by adding safety margins to a circuit's operating voltage, clock frequency or charge stored per bit. However, the rising energy costs needed to compensate for increasing unpredictability are rapidly becoming unacceptable in today's environment where power consumption is often the limiting factor on integrated circuit performance and energy efficiency is a national concern. Reliability and energy consumption are both reaching key inflection points that, together, threaten to reduce or end the benefits of feature size reduction. To continue beneficial scaling, we must use a cross-layer, Jull-system-design approach to reliability. Unlike current systems, which charge every device a substantial energy tax in order to guarantee correct operation in spite of rare events, such as one high-threshold transistor in a billion or one erroneous gate evaluation in an hour of computation, cross-layer reliability schemes make reliability management a cooperative effort across the system stack, sharing information across layers so that they only expend energy on reliability when an error actually occurs. Figure 1 illustrates an example of such a system that uses a combination of information from the application and cheap architecture-level techniques to detect errors. When an error occurs, mechanisms at higher levels in the stack correct the error, efficiently delivering correct operation to the user in spite of errors at the device or circuit levels. In the realms of memory and communication, engineers have a long history of success in tolerating unpredictable effects such as fabrication variability, transient upsets, and lifetime wear using information sharing, limited redundancy, and cross-layer approaches that anticipate, accommodate, and suppress errors. Networks use a combination of hardware and software to guarantee end-toend correctness. Error-detection and correction codes use additional information to correct the most common errors, single-bit transmission errors. When errors occur that cannot be corrected by these codes, the network protocol requests re-transmission of one or more packets until the correct data is received. Similarly, computer memory systems exploit a cross-layer division of labor to achieve high performance with modest hardware. Rather than demanding that hardware alone provide the virtual memory abstraction, software page-fault and TLB-miss handlers allow a modest piece of hardware, the TLB, to handle the common-case operations on a cyc1e-by-cycle basis while infrequent misses are handled in system software. Unfortunately, mitigating logic errors is not as simple or as well researched as memory or communication systems. This lack of understanding has led to very expensive solutions. For example, triple-modular redundancy masks errors by triplicating computations in either time or area. This mitigation methods imposes a 200% increase in energy consumption for every operation, not just the uncommon failure cases. At a time when computation is rapidly becoming part of our critical civilian and military infrastructure and decreasing costsfor computation are fueling our economy and our well being, we cannot afford increasingly unreliable electronics or a stagnation in capabilities per dollar, watt, or cubic meter. If researchers are able to develop techniques that tolerate the growing unpredictability of silicon devices, Moore's Law scaling should continue until at least 2022. During this 12-year time period, transistors, which are the building blocks of electronic devices, will scale their dimensions (feature sizes) from 45nm to 4.5nm.