We are accustomed to thinking of computers as fail-stop, especially the cores that execute instructions, and most system software implicitly relies on that assumption. During most of the VLSI era, processors that passed manufacturing tests and were operated within specifications have insulated us from this fiction. As fabrication pushes towards smaller feature sizes and more elaborate computational structures, and as increasingly specialized instruction-silicon pairings are introduced to improve performance, we have observed ephemeral computational errors that were not detected during manufacturing tests. These defects cannot always be mitigated by techniques such as microcode updates, and may be correlated to specific components within the processor, allowing small code changes to effect large shifts in reliability. Worse, these failures are often "silent" - the only symptom is an erroneous computation.

Spanner is Google’s scalable, multiversion, globally distributed, and synchronously replicated database. It is the first system to distribute data at global scale and support externally-consistent distributed transactions. This article describes how Spanner is structured, its feature set, the rationale underlying various design decisions, and a novel time API that exposes clock uncertainty. This API and its implementation are critical to supporting external consistency and a variety of powerful features: nonblocking reads in the past, lock-free snapshot transactions, and atomic schema changes, across all of Spanner.

This article introduces MPI-F an efficient implementation of MPI on the IBM-SP1 distributed memory cluster. After discussing the novel and key concepts of MPI and how they relate to an implementation, the MPI-F system architecture is outlined in detail. Although many incorrectly assume that MPI will not be efficient due to its increased functionality, MPI-F performance demonstrates efficiency as good as the best message passing library currently available on the SP1.

Revue des methodes de coupures multiples appliquees a la determination de la complexite des circuits VLSI, basees sur une definition generalisee du «contenu d'information». Les techniques presentees conviennent aux problemes autorisant ou non la repetition des entrees. On demontre que la complexite des «codages redondants» dans le calcul n'est pas reduite par la repetition des entrees. On deduit une borne de complexite. Les deux parametres principaux du calcul des circuits VLSI sont la superficie et le temps

We introduce new paradigms for the construction of efficient parallel graph algorithms. These paradigms, called filtration and funnelled pipelining, are illustrated with VLSI circuits for computing connected components, minimum spanning forests, and biconnected components. These circuits use realistic I/O schedules and require time and area of O(n1+ε). Thus they are essentially optimal. Filtration is a technique used to rapidly discard irrelevant input data. This greatly reduces storage, time, and communications costs in a wide variety of problems. A funnelled pipeline is obtained by building a series of increasingly thorough filter stages. Transition times along such a pipeline of filters form an exponentially increasing sequence. The increasing amount of time exactly balances the increasing degree of filtration. This balance makes possible the cascaded filtration critical to the minimum spanning forest and the biconnected components algorithms.

Dynamic programming is used in speech recognition to search efficiently for word sequences whose templates best match acoustic data. The search is constrained by finite-state networks embodying grammatical rules. Typically, dynamic programming is implemented in two steps: the first calculates, for each state in a network and for each time, the best way of arriving at that state at that time; the second traces back from the final state at the final time to the initial state at the initial time to determine the best path through the network. This second step cannot be initiated before the determination (usually from the detection of silence) that the final state has been reached. Such a determination is difficult in the recognition of truly continuous speech; there are often no reliable anchor points. Further, it is often desirable to be able to recognize at least part of an utterance before a speaker has stopped talking. In this paper we introduce a technique for discovering the initial section of the optimal path through a network before the traversal of the network is complete. It can be used to report a system's interpretation of acoustic data from the not-too-distant past without relying on or making any decisions which may degrade recognition accuracy.