Analyses and Optimizations for Shared Address Space Programs

We present compiler analyses and optimizations for explicitly parallel programs that communicate through a shared address space. Any type of code motion on explicitly parallel programs requires a new kind of analysis to ensure that operations reordered on one processor cannot be observed by another. The analysis, calledcycle detection, is based on work by Shasha and Snir and checks for cycles among interfering accesses. We improve the accuracy of their analysis by using additional information fromsynchronization analysis, which handles post–wait synchronization, barriers, and locks. We also make the analysis efficient by exploiting the common code image property of SPMD programs. We make no assumptions on the use of synchronization constructs: our transformations preserve program meaning even in the presence of race conditions, user-defined spin locks, or other synchronization mechanisms built from shared memory. However, programs that use linguistic synchronization constructs rather than their user-defined shared memory counterparts will benefit from more accurate analysis and therefore better optimization. We demonstrate the use of this analysis for communication optimizations on distributed memory machines by automatically transforming programs written in a conventional shared memory style into a Split-C program, which has primitives for nonblocking memory operations and one-way communication. The optimizations includemessage pipelining, to allow multiple outstanding remote memory operations, conversion of two-way to one-way communication, and elimination of communication through data reuse. The performance improvements are as high as 20–35% for programs running on a CM-5 multiprocessor using the Split-C language as a global address layer. Even larger benefits can be expected on machines with higher communication latency relative to processor speed.     

Chasing shared understanding in drilling operations

Lack of shared understanding is frequently found to be the main cause when accidents are investigated. Still, few studies explicitly explore and document the causal effects of shared understanding in successful work. Thus, the attribution of insufficient shared understanding as an accident cause lacks the substantiation of shared understanding as a contributor to successful work. In this article a case of measurement discrepancies in an offshore drilling operation is studied, and in the elaboration of the case shared understanding is found not to qualify as a condition with significant impact on the collaborative work. One important reason for this is the epistemological inadequacy of the different concepts of shared understanding. Although more critical research on shared understanding is needed before one can conclude more generic on this topic, the findings are important to the current development of Integrated Operations where shared understanding is pointed out as an important target area.     

A Timing Assumption and Two t-Resilient Protocols for Implementing an Eventual Leader Service in Asynchronous Shared Memory Systems

This paper considers the problem of electing an eventual leader in an asynchronous shared memory system. While this problem has received a lot of attention in message-passing systems, very few solutions have been proposed for shared memory systems. As an eventual leader cannot be elected in a pure asynchronous system prone to process crashes, the paper first proposes to enrich the asynchronous system model with an additional assumption. That assumption (denoted AWB) is particularly weak. It is made up of two complementary parts. More precisely, it requires that, after some time, (1) there is a process whose write accesses to some shared variables be timely, and (2) the timers of (t−f) other processes be asymptotically well-behaved (t denotes the maximal number of processes that may crash, and f the actual number of process crashes in a run). The asymptotically well-behaved timer notion is a new notion that generalizes and weakens the traditional notion of timers whose durations are required to monotonically increase when the values they are set to increase (a timer works incorrectly when it expires at arbitrary times, i.e., independently of the value it has been set to). The paper then focuses on the design of t-resilient AWB-based eventual leader protocols. “t-resilient” means that each protocol can cope with up to t process crashes (taking t=n−1 provides wait-free protocols, i.e., protocols that can cope with any number of process failures). Two protocols are presented. The first enjoys the following noteworthy properties: after some time only the elected leader has to write the shared memory, and all but one shared variables have a bounded domain, be the execution finite or infinite. This protocol is consequently optimal with respect to the number of processes that have to write the shared memory. The second protocol guarantees that all the shared variables have a bounded domain. This is obtained at the following additional price: t+1 processes are required to forever write the shared memory. A theorem is proved which states that this price has to be paid by any protocol that elects an eventual leader in a bounded shared memory model. This second protocol is consequently optimal with respect to the number of processes that have to write in such a constrained memory model. In a very interesting way, these protocols show an inherent tradeoff relating the number of processes that have to write the shared memory and the bounded/unbounded attribute of that memory.     

A secure distributed framework for achieving k-anonymity

k-anonymity provides a measure of privacy protection by preventing re-identification of data to fewer than a group of k data items. While algorithms exist for producing k-anonymous data, the model has been that of a single source wanting to publish data. Due to privacy issues, it is common that data from different sites cannot be shared directly. Therefore, this paper presents a two-party framework along with an application that generates k-anonymous data from two vertically partitioned sources without disclosing data from one site to the other. The framework is privacy preserving in the sense that it satisfies the secure definition commonly defined in the literature of Secure Multiparty Computation.     

Computing in Totally Anonymous Asynchronous Shared Memory Systems

In the totally anonymous shared memory model of asynchronous distributed computing, processes have no identifiers and run identical programs. Moreover, processes have identical interface to the shared memory, and in particular, there are no single-writer registers. This paper assumes that processes do not fail, and the shared memory consists only of read/write registers, which are initialized to some default value. A complete characterization of the functions and agreement tasks that can be solved in this model is presented. Furthermore, it is shown that if a function is computable, then two registers are sufficient for some algorithm to compute it. Consensus is an important agreement task that can be computed. The paper proves logarithmic lower bounds on the number of registers and rounds needed for solving consensus in this model. A consensus protocol using a linear number of shared registers and rounds is also presented.     

Predicting the Future of Discrete Sequences from Fractal Representations of the Past

We propose a novel approach for building finite memory predictive models similar in spirit to variable memory length Markov models (VLMMs). The models are constructed by first transforming the n-block structure of the training sequence into a geometric structure of points in a unit hypercube, such that the longer is the common suffix shared by any two n-blocks, the closer lie their point representations. Such a transformation embodies a Markov assumption—n-blocks with long common suffixes are likely to produce similar continuations. Prediction contexts are found by detecting clusters in the geometric n-block representation of the training sequence via vector quantization. We compare our model with both the classical (fixed order) and variable memory length Markov models on five data sets with different memory and stochastic components. Fixed order Markov models (MMs) fail on three large data sets on which the advantage of allowing variable memory length can be exploited. On these data sets, our predictive models have a superior, or comparable performance to that of VLMMs, yet, their construction is fully automatic, which, is shown to be problematic in the case of VLMMs. On one data set, VLMMs are outperformed by the classical MMs. On this set, our models perform significantly better than MMs. On the remaining data set, classical MMs outperform the variable context length strategies.     

Product platform design through sensitivity analysis and cluster analysis

Scale-based product platform design consists of platform configuration to decide which variables are shared among which product variants, and selection of the optimal values for platform (shared) and non-platform variables for all product variants. The configuration step plays a vital role in determining two important aspects of a product family: efficiency (cost savings due to commonality) and effectiveness (capability to satisfy performance requirements). Many existing product platform design methods ignore it, assuming a given platform configuration. Most approaches, whether or not they consider the configuration step, are single-platform methods, in which design variables are either shared across all product variants or not shared at all. In multiple-platform design, design variables may be shared among variants in any possible combination of subsets, offering opportunities for superior overall design but presenting a more difficult computational problem. In this work, sensitivity analysis and cluster analysis are used to improve both efficiency and effectiveness of a scale-based product family through multiple-platform product family design. Sensitivity analysis is performed on each design variable to help select candidate platform design variables and to provide guidance for cluster analysis. Cluster analysis, using performance loss due to commonization as the clustering criterion, is employed to determine platform configuration. An illustrative example is used to demonstrate the merits of the proposed method, and the results are compared with existing results from the literature.     

Shared models: The cognitive equivalent of aLingua Franca

The richness of humanity is the diversity of its cultures, but now as never before the destructive power of modern technology and threatening ecological disasters make it necessary that we all recognize we are many peoples of one world. Complementing the diversity of our different cultures, the growth of a common, scientific knowledge inspires the hope that we may achieve and share a secondary culture of ideas. Computers, which can help represent explicitly the best ideas of modern science, can aid in the diffusion of such powerful ideas to create a popular, secondary, scientific culture.We propose that a primary objective of learning environment design should be the development of thinkable models in the minds of students; further, since thinkable models represent a level of knowledge deeper than superficial terminology and specific graphical forms, such may be an objective suitable for various people in different cultures. To pursue these notions in some detail, a taxonomy of models is developed and the issue of how representations relate to human modes of perception and action is raised. The notions are explored first through the contrasting of a half-dozen approaches to the Pythagorean Theorem; then through describing polylingual word worlds and the Rosetta disk project.Common experiences produce common models —not only public models but the cognitive structures built from those shared experiences; and shared models will permit enhanced communication and understanding. If sharing experiences through play with computer based learning environments in different languages permits children to develop common models of the world, this will ultimately enhance mutual understanding between people in different places. This objective is now within reach.     

Combining flow and dependence analyses to expose redundant array accesses

The success of large-scale, hierarchical and distributed shared memory systems hinges on our ability to reduce delays resulting from remote accesses to shared data. To facilitate this, we present a compile-time algorithm for analyzing programs with doall-style parallelism to determine when read and write accesses to shared data areredundant (unnecessary). One identified, redundant remote accesses can be replaced by local accesses or eliminated entirely. This optimization improves program performance in two ways. First, slow memory accesses are replaced by faster ones. Second, the time to perform other remote memory accesses may be reduced as a result of the decreased traffic level. We also show how the information obtained through redundancy analysis can be used for other compiler optimizations such as prefetching and cache management.     

Remote collaboration on desk‐sized displays

Work on remote collaboration has often focused on the person space created by a conventional videoconference where the participants see each other's faces, but we argue that a task space containing shared visual information is more important for most tasks. Trends in display technology mean that large visual task spaces can be created to maximize the shared context between collaborators. We have created a system called the Escritoire that presents users with a desk‐sized projected display with bimanual input that allows documents and images to be arranged and modified by multiple remote collaborators. We describe the software architecture, the protocol that is used between the client and server programs, and the pen traces we have added to allow participants to gesture to each other in the large visual space to enrich their communication. Our user trials have shown that participants were able to use the system with a minimum of training, and found the traces useful in the collaborative setting. In future we will be connecting three or more sites together to explore the issues that arise with multi‐party interaction on large shared desks. Copyright © 2005 John Wiley & Sons, Ltd.     

V_THR: An Adaptive Load Balancing Algorithm

This paper presents a new adaptive algorithm for dynamic load balancing on a shared BUS architecture. We present results obtained from simulation studies and queuing analysis, which reflect the relation between the BUS contention and the efficiency of load balancing. The proposed algorithm uses a scheme for monitoring theThresholdparameter to dynamically adapt itself to the limited bandwidth of the shared BUS. We have compared our algorithm with some of the known policies, and the relative performance appears to be promising.     

Dynamic Affinity Cluster Allocation in a Shared Disks Cluster

A shared disks (SD) cluster couples multiple computing nodes for high performance transaction processing, and all nodes share a common database at the disk level. In the SD cluster, a front-end router selects a node for an incoming transaction to be executed. An affinity-based routing can increase the buffer hit ratio of each node by clustering transactions referencing similar data to be executed on the same node. However, the affinity-based routing is non-adaptive to the changes of the system load. This means that a specific node would be overloaded if corresponding transactions rush into the system. In this paper, we propose a new transaction routing algorithm, named Dynamic Affinity Cluster Allocation (DACA). DACA can make an optimal balance between the affinity-based routing and indiscriminate sharing of load in the SD cluster. As a result, DACA can increase the buffer hit ratio and reduce the frequency of inter-node buffer invalidations while achieving the dynamic load balancing.     

An implementation of distributed shared memory

Shared memory is a simple yet powerful paradigm for structuring systems. Recently, there has been an interest in extending this paradigm to non-shared memory architectures as well. For example, the virtual address spaces for all objects in a distributed object-based system could be viewed as constituting a global distributed shared memory. We propose a set of primitives for managing distributed shared memory. We present an implementation of these primitives in the context of an object-based operating system as well as on top of Unix.     

GeoQuorums: implementing atomic memory in mobile ad hoc networks

We present a new approach, the GeoQuorums approach, for implementing atomic read/write shared memory in mobile ad hoc networks. Our approach is based on associating abstract atomic objects with certain geographic locations. We assume the existence of focal points, geographic areas that are normally “populated” by mobile nodes. For example, a focal point may be a road junction, a scenic observation point, or a water resource in the desert. Mobile nodes that happen to populate a focal point participate in implementing a shared atomic object, using a replicated state machine approach. These objects, which we call focal point objects, are prone to occasional failures when the corresponding geographic areas are depopulated. The GeoQuorums algorithm uses the fault-prone focal point objects to implement atomic read/write operations on a fault-tolerant virtual shared object. The GeoQuorums algorithm uses a quorum-based strategy in which each quorum consists of a set of focal point objects. The quorums are used to maintain the consistency of the shared memory and to tolerate limited failures of the focal point objects, which may be caused by depopulation of the corresponding geographic areas. We present a mechanism for changing the set of quorums on the fly, thus improving efficiency. Overall, the new GeoQuorums algorithm efficiently implements read and write operations in a highly dynamic, mobile network.     

The butterfly barrier

We describe and algorithm for barrier synchronization that requires only read and write to shared store. The algorithm is faster than the traditionallocked counter approach for two processors and has an attractive log₂ N time scaling for largerN. The algorithm is free of hot spots and critical regions and requires a shared memory bandwidth which grows linearly withN, the number of participating processors. We verify the technique using both a real shared memory multiprocessor, for numbers of processors up to 30, and a shared memory multiprocessor simulator, for number of processors up to 256.Work performed under the auspices of the U.S. Department of Energy by the Lawrence Livermore National Laboratory under contract No. W-7405-ENG-48.     

Combining classifiers: A theoretical framework

The problem of classifier combination is considered in the context of the two main fusion scenarios: fusion of opinions based on identical and on distinct representations. We develop a theoretical framework for classifier combination for these two scenarios. For multiple experts using distinct representations we argue that many existing schemes such as the product rule, sum rule, min rule, max rule, majority voting, and weighted combination, can be considered as special cases of compound classification. We then consider the effect of classifier combination in the case of multiple experts using a shared representation where the aim of fusion is to obtain a better estimate of the appropriatea posteriori class probabilities. We also show that the two theoretical frameworks can be used for devising fusion strategies when the individual experts use features some of which are shared and the remaining ones distinct. We show that in both cases (distinct and shared representations), the expert fusion involves the computation of a linear or nonlinear function of thea posteriori class probabilities estimated by the individual experts. Classifier combination can therefore be viewed as a multistage classification process whereby thea posteriori class probabilities generated by the individual classifiers are considered as features for a second stage classification scheme. Most importantly, when the linear or nonlinear combination functions are obtained by training, the distinctions between the two scenarios fade away, and one can view classifier fusion in a unified way.     

Do You Know What I Know? A Shared Understandings Perspective on Text‐Based Communication

This article illustrates how the hermeneutic analysis of text illuminates how shared understandings affect our interpretations of lean communication in distributed work environments. It is proposed that in contrast to the pessimistic conclusions of media richness theory that lean communication channels cannot support complex or equivocal work tasks, miscommunications are not the result of technology, but rather occur due to a lack of shared understandings among the individuals communicating. An illustrative case study based on fieldwork in franchise organizations is presented to demonstrate the possibilities for how the hermeneutic analysis of coherence, invention, intention, and reference can be used to discover how workers create and recreate shared understandings through text.     

A Compiler-Directed Approach to Network Latency Reduction for Distributed Shared Memory Multiprocessors

In distributed shared memory multiprocessor systems, parallel tasks communicate through sharing memory data. As the system size increases, such communication cost becomes the main factor that limits the overall parallelism and performance. In this paper, we propose a new solution to the problem through judiciously managing the relevant resource, namely, the shared data and the interconnection network (IN) through which the sharing is carried out. In this approach, communication cost is minimized by means of data migration/allocation which is based on analyzing general layered task graphs, sharing behavior of parallel tasks, and network topology. Our method is not applicable for read only variables. Further, for the time being, the usefulness of the method is limited to multiprocessors where no cache coherence mechanism is implemented. Four typical interconnection topologies for multiprocessors are considered, namely, shared-bus, hierarchical-bus, 2-D mesh, and fat-tree structures. Efficient data allocation algorithms for each of the four network topologies are developed that make decision on data allocation/migration at the compile time. The complexity of one algorithm isO(np) for shared-bus andO(n²p) for the remaining three in a system withnprocessors executing ap-layer task graph for one shared variable. We have also given an algorithm to determine optimal allocation/migration scheme for multiple shared variables. However, the cost of the algorithm become prohibitive when the number of shared variables is high. Therefore, a heuristic of low complexity is suggested. The heuristic is optimal for some topologies.     

Uniting Political Bloggers in Diversity: Collective Identity and Web Activism

In contrast to the common view of blogging as a highly narcissistic activity, this study explicates how blogging is a communal activity and the emergence of a collective identity which drives collective action. 41 interviews with activist and nonactivist bloggers revealed that shared consciousness, distinctive blogging practices, and the articulation of a common adversary contribute to the development of a collective identity. Furthermore, identity multiplexity points to the emergence of “individualized collectiveness,” which extends beyond networked individualism and is reinforced by offline participation in activism and pre‐existing social ties with other activists.     

A case study of co-ordinative decision-making in disaster management

A persistent problem in the management of response to disasters is the lack of coordination between the various agencies involved. This paper reports a case study of inter-agency co-ordination during the response to a railway accident in the UK. The case study examined two potential sources of difficulty for coordination: first, poorly shared mental models; and, second, a possible conflict between the requirements of distributed decision-making and the nature of individual decision-making. Interviews were conducted with six individuals from three response agencies. Analysis of reported events suggested that inter-agency co-ordination suffered through a widespread difficulty in constructing a reflexive shared mental model; that is, a shared mental representation of the distributed decision-making process itself, and its participants. This difficulty may be an inherent problem in the flexible development of temporary multi-agency organizations. The analysis focused on a distributed decision over how to transport casualties from an isolated location to hospital. This decision invoked a technique identified here as the progression of multiple options, which contrasts with both recognition-primed and analytical models of individual decisionmaking. The progression of multiple options appeared to be an effective technique for dealing with uncertainty, but was a further source of difficulty for inter-agency co-ordination.