Look-ahead Mechanism Research Articles

Parallel Job Scheduling in Cloud with Lookahead and Workload Consolidation

The cloud computing paradigm enables consumers to run their applications in remote data centers. Many of these applications may be complex which requires parallel processing capabilities. Parallel job scheduling techniques mainly focus on improving responsiveness and utilization. For a data center that deals with parallel jobs, it is important to devise an optimal schedule which results in maximal utilization of available node capacity. For that, this paper propose a parallel job scheduling technique which uses the key concepts such as workload consolidation through virtualization technologies and backfilling with look ahead mechanism. The proposed method is compared to scheduling using backfilling technique with workload consolidation. The results show that the proposed method with lookahead mechanism has shown better performance. General Terms Cloud computing, Parallel Job Scheduling, Backfilling, Resource Consolidation.

A Parthenogenetic Algorithm for the Founder Sequence Reconstruction Problem

The maximum fragment length (MFL) is an important computational model for solving the founder sequence reconstruction problem. Benedettini et al. presented a meta-heuristic algorithm BACKFORTH based on iterative greedy method. The BACKFORTH algorithm starts with a single initial solution, and iteratively alternates between a partial destruction and reconstruction in order to obtain a final solution. The kind of optimization mechanism, which is based on a single initial solution, may make the performance of the BACKFORTH algorithm sensitive to the quality of the initialization. In this paper, a practical parthenogenetic algorithm PGMFL, which is a population-based meta-heuristic method, is proposed. The PGMFL algorithm can search multiple regions of a solution space simultaneously. A novel genetic operator is introduced based on the presented heuristic algorithm HF, which takes advantage of look-ahead mechanism and some potential information, i.e., the proportions of 0 and 1 entries in a column of recombinant matrix and those in the corresponding column of the founder matrix, and some other heuristic information, to compute the column values. The PGMFL algorithm can get fewer breakpoints and longer fragment average length than the BACKFORTH algorithm, which are proved by a number of experiments.

Hybrid solution method for resource-constrained project scheduling problem using a new schedule generator

A number of exact scheduling schemes incorporating various solution procedures such as branch and bound and dynamic programming exist for solving the well-known resource-constrained project scheduling problem. In this paper, we report on an efficient hybrid method which prominently depends on a branch and bound with a look-ahead mechanism as well as a genetic algorithm coupled with a new schedule generation scheme called tri-directional schedule generator. Minimal delay set, core time, and left shift are among the pruning rules used to prune inferior nodes of the enumeration tree. The proposed method has been verified and validated using a standard set of test problems with 30–120 activities requiring between one and six resource types each. The capability and applicability of the proposed method are demonstrated using standard problem instances.

A hybrid Ant algorithm for the set covering problem

Set covering problem is the model for many important industrial applications. In this paper, we solve some benchmarks of this problem with ant colony optimization algorithms using a new transition rule. A look-ahead mechanism was incorporated to check constraint consistency in ant computing. Computational results are presented showing the advantages to use this additional mechanism to ant system and ant colony system. Key words: Set covering problem, ant colony optimization, look-ahead techniques.

𝒞𝒪𝒞𝒪𝒩ℱ: Conceptual language-based configuration

This article deals with the configuration task. Configuring means assembling a set of predefined components in order to build a composite object that meets a set of requirements. Here we present COCONF, an approach to configuration based on a conceptual encoding of the configuration knowledge, directly exploitable by a software configuration system to compute configurations. In particular, we show how Conflict-Directed Backjumping can be adapted to the proposed framework and in which way the efficiency of the configurator may be still enhanced by two look-ahead mechanisms, which exploit the characteristics of the modeling language and the explicit representation of both the compositional structure and the taxonomic relations among component types. The configuration algorithms are explained in detail; the assumptions and the properties which they are based on are explicitly stated. Formal proofs are provided for the basic properties. A set of experimental results on three different real-world domains are presented, which prove the suitability of the approach.

A lattice search technique for a long-contextual-span hidden trajectory model of speech

We have recently developed a long-contextual-span hidden trajectory model (HTM) which captures underlying dynamic structure of speech coarticulation and reduction. Due to the long-span nature of the HTM and the complexity of its likelihood score computation, N-best list rescoring was the principal paradigm for evaluating the HTM for phonetic recognition in our earlier work. In this paper, we describe improved likelihood score computation in the HTM and a novel A ∗-based time-asynchronous lattice-constrained decoding algorithm for the HTM evaluation. We focus on several special considerations in the decoder design, which are necessitated by the dependency of the HTM score at each given frame on the model parameters associated with a variable number of adjacent past and future phones. We present details on how the nodes and links in the lattices are expanded via a look-ahead mechanism, on how the A ∗ heuristics are estimated, and on how pruning strategies are applied to speed up the search process. The experiments on the standard TIMIT phonetic recognition task show improvement of recognition accuracy by the new search algorithm on recognition lattices over the traditional N-best rescoring paradigm.

Highly responsive and efficient QoS routing using pre- and on-demand computations along with a new normal measure

Multi-constrained path (MCP) selection is one of the great challenges that QoS routing (QoSR) faces. To address it in an efficient and highly responsive manner, we propose a new QoSR algorithm, namely NM_MCP (normal measure-based multiple constrained path). Using the Dijkstra's algorithm with respect to each link metric, NM_MCP pre-computes k primary paths in advance, where k is the number of link weights. When a routing request arrives, NM_MCP executes a modified version of the Dijkstra's algorithm using a newly proposed, normal-measure-based nonlinear cost function. Extensive simulations show that NM_MCP achieves higher success rate in finding feasible paths with less computational cost than existing algorithms. To further improve the performance, we incorporate Pareto and nonlinear look-ahead mechanisms into the algorithm.

Dynamic lot sizing and tool management in automated manufacturing systems

The overall aim of this study is to show that there is a critical interface between the lot sizing and tool management decisions, and these two problems cannot be viewed in isolation. We propose five alternative algorithms to solve lot sizing, tool allocation and machining conditions optimization problems simultaneously. The first algorithm is an exact algorithm which finds the global optimum solution, and the others are heuristics equipped with a look-ahead mechanism to guarantee at least local optimality. The computational results indicate that the amount of improvement is statistically significant for a set of randomly generated problems. The magnitude of cost savings is dependent on the system parameters. Scope and purpose In most of the studies on tool management, lot sizes are taken as a predetermined input while deciding on tool allocations and machining parameters. This might create empty feasible solution spaces and otherwise unnecessarily limit the number of alternatives possible for the tool management problem. In this study, we consider the integration of lot sizing and tool management problems to minimize total production cost for multiple periods under dynamic demand. By integrating these decisions we not only improve the overall solution, but also prevent any infeasibility that might occur for the tool management problem due to decisions made at the lot sizing level.

Load-balancing data prefetching techniques

Despite the success of hybrid data address and value prediction in increasing the accuracy and coverage of data prefetching, memory access latency is still found to be an important bottleneck to the system performance. Careful study shows that about half of the cache misses are actually due to data references whose access pattern can be predicted accurately. Furthermore, the overall cache effectiveness is bounded by the behavior of unpredictable data references in cache. In this paper, we propose a set of four load-balancing techniques to address this memory latency problem. The first two mechanisms, sequential unification and aggressive lookahead mechanisms, are mainly used to reduce the chance of partial hits and the abortion of accurate prefetch requests. The latter two mechanisms, default prefetching and cache partitioning mechanisms, are used to optimize the cache performance of unpredictable references. The resulting cache, called the LBD ( load-balancing data) cache, is found to have superior performance over a wide range of applications. Simulation of the LBD cache with RPT prefetching (reference prediction table — one of the most cited selective data prefetch schemes proposed by Chen and Baer) on SPEC95 showed that significant reduction in data reference latency, ranging from about 20 to over 90% and with an average of 55.89%, can be obtained. This is compared against the performance of prefetch-on-miss and RPT, with an average latency reduction of only 17.37 and 26.05%, respectively.

Modelling And Distributed Simulation On A Network Of Workstations

This paper addresses the issues of applying parallel simulation techniques to accelerate simulation on a network of workstations. The Omega multistage interconnection network is used as an example application. The conservative parallel simulation synchronization scheme using conventional null-message approach to resolve deadlock problem is based on a lookahead mechanism. For some application domains, unfortunately, the lookahead information is not available. Consequently, parallel simulation using null messages can result in live lock. We propose a deadlock/livelock free scheme using null messages, but without the lookahead assumption, to coordinate the simulation. In addition, we investigate different partitioning and transformation techniques for mapping a simulation program onto a network of workstations. A flushing mechanism to address the combinatoric explosion of using null-message in conservative simulation is also discussed. Our analysis shows that the proposed flushing mechanism effectively reduces the number of null messages from exponential to linear.

DEALING WITH AMBIGUITIES IN ENGLISH CONJUNCTIONS AND COMPARATIVES BY A DETERMINISTIC PARSER

The major problems in parsing English conjunctions and comparatives are ambiguities of scoping and ellipsis. Scoping ambiguities occur when a parser cannot deterministically detect boundaries of constituents, while ellipsis ambiguities occur when a parser cannot deterministically detect missing components. Since simple lookahead mechanisms cannot collect adequate information to resolve these ambiguities, a parsing strategy that only employs such mechanisms will need to backtrack each time it makes incorrect assumptions. In this paper, we extend the Wait-And-See strategy to parse conjunctions and comparatives deterministically and simultaneously. Several mechanisms, such as bottom-up preparsing, suspension, and pattern matching, are implemented. The bottom-up preparsing accesses the dictionary and recognizes isolated sentence fragments which can be determined without ambiguities. The suspension, which is different from Marcus’s attention shifting, allows the parser to suspend temporally at ambiguous points and continue to parse the rest of the sentence until it obtains the necessary information to resolve the ambiguities. Pattern matching uses the concept of symmetry to detect missing components (the ellipses) in the two conjoined or compared sentence fragments.

The effects of antagonistic gestures on temporal and amplitude parameters of anticipatory labial coarticulation.

Anticipatory lip rounding before neutral versus antagonistic vowel environments was studied in a test of the compatibility notion of a look-ahead scanning mechanism (Henke, 1967). Electromyograms were obtained from orbicularis oris superior, an agonist for lip rounding, and risorius, an agonist for lip spreading, from three normal speakers repeating the nonsense disyllables /tiku/, /taku/, /tuki/, /kikstu/, and /kakstu/ twenty times each. When the syllable-initial vowel was the lip spread vowel /i/, oris activity began earlier and with greater force than for syllable-initial /a/ contexts across all speakers. This finding argues for a revision of a key assumption underlying the look-ahead scan model for anticipatory lip rounding. The encoding program for speech exhibits temporal and amplitude adjustments in its coarticulatory behavior to accommodate in an active way contradictory neuromuscular and biomechanical conditions. These results are discussed with reference to current models of speech production.

Electromyographic study of the velum during speech

Electromyographic recordings of the levator palatini were obtained to investigate the relationship between velar movement and its motor command. Two Japanese subjects read a list of meaningful disyllabic test words. There was no systematic difference in EMG level between either voiced and voiceless or stop and fricative consonants. Rather, the activity level for a given nonnasal consonant varies with its phonetic environment. Electromyographically, anticipatory nasal coarticulation generally followed the so-called look-ahead mechanism, although in some examples there seemed to be a restriction of the anticipatory effect. On the other hand, carry-over nasal coarticulation was not as pervasive as the anticipatory effect, as the carry-over effect was present for vowel segments following syllable-initial nasals, while it was absent for vowel segments following syllable-final nasals.

Electromyographic study of velar movements during speech

Electromyographic recording of the levator palatini was performed to investigate the relationship between velar movements and the motor command. Two Japanese subjects read a list of meaningful disyllabic test words. There was no systematic difference in EMG level between either voiced and voiceless or stop and fricative consonants. Rather, the activity level for a certain non-nasal consonant appeared to vary according to the phonetic environment. Our EMG data seemed to be consistent with our previous fiberoptic findings on the difference in the degree of nasalization between syllable-initial and syllable-final nasals [Ushijima and Sawashima, Annual Bull. RILP Univ. Tokyo 6, 25 (1972)]. Electromyographically, the right-to-left effect of nasalization followed the so-called look-ahead mechanism, although in some examples there seemed to be a restriction of anticipatory coarticulation effect. On the other hand, the left-to-right effect manifested itself as increasing levator activity for postnasal vowel segments. The findings suggest that there exists a different mechanism between the right-to-left and left-to-right effects of coarticulation at the level of motor command. [Supported by NIDR Grant DE-1774.]

Feedback coupled resource allocation policies in the multiprogramming-multiprocessor computer system

This paper presents model studies of some integrated feedback-driven scheduling systems for a multiprogrammed computer system. This abstract can present only the conclusions of the studies and little of the supporting data and detail. The basic format of the analysis is to fix a size for the local buffers and a total size for the collection buffers, to define a set of algorithms for the determination of the data removal quanta to the local buffers, the allocation of space in the collection buffers, and the look-ahead mechanism for input and then to evaluate the relative merits of the various strategies by the resulting CPU efficiency. Three feedback algorithms are studied as examples in this work.