Search

Lambda paradigm, also known as Function as a Service (FaaS), is a novel event-driven concept that allows companies to build scalable and reliable enterprise applications in an off-premise computing data-center as a serverless solution. In practice, however, an important goal for the service provider of a Lambda platform is to devise an efficient way to consolidate multiple Lambda functions in a single host. While the majority of existing resource management solutions use only operating-system level metrics (e.g., average utilization of computing and I/O resources) to allocate the available resources among the submitted workloads in a balanced way, a resource allocation schema that is oblivious to the issue of shared-resource contention can result in a significant performance variability and degradation within the entire platform. This paper proposes a predictive controller scheme that dynamically allocates resources in a Lambda platform. This scheme uses a prediction tool to estimate the future rate of every event stream and takes into account the quality of service enforcements requested by the owner of each Lambda function. This is formulated as an optimization problem where a set of cost functions are introduced (i) to reduce the total QoS violation incidents; (ii) to keep the CPU utilization level within an accepted range; and (iii) to avoid the fierce contention among collocated applications for obtaining shared resources. Performance evaluation is carried out by comparing the proposed solution with an enhanced interference-aware version of three well-known heuristics, namely spread, binpack (the two native clustering solutions employed by Docker Swarm) and best-effort resource allocation schema. Experimental results show that the proposed controller improves the overall performance (in terms of reducing the end-to-end response time) by 14.9 percent on average compared to the best result of the other heuristics. The proposed solution also increases the overall CPU utilization by 18 percent on average (for lightweight workloads), while achieves an average 87 percent (maximum 146 percent) improvement in preventing QoS violation incidents. [ABSTRACT FROM AUTHOR]

Virtualized Data Centers are packed with numerous web and cloud services nowadays. In such large infrastructures, providing reliable service platforms depends heavily on efficient sharing of physical machines (PMs) by virtual machines (VMs). To achieve efficient consolidation, performance degradation of co-located VMs must be correctly understood, modeled, and predicted. This work is a major step toward understanding such baffling phenomena by not only identifying, but also quantifying sensitivity of general purpose VMs to their demanded resources. vmBBProfiler, our proposed system in this work, is able to systematically profile behavior of any general purpose VM and calculate its sensitivity to system provided resources such as CPU, Memory, and Disk. vmBBProfiler is evaluated using 12 well-known benchmarks, varying from pure CPU/Mem/Disk VMs to mixtures of them, on three different PMs in our VMware-vSphere based private cloud. Extensive empirical results conducted over 1200 h of profiling prove the efficiency of our proposed models and solutions; it also opens doors for further research in this area. [ABSTRACT FROM AUTHOR]

This paper first envisions the advent of the wireless mesh networks with multiple radios and directional antennas in future. Then, based on the observation that simplicity induces efficiency and scalability, the paper proposes a joint geometric deployment and routing strategy for such mesh networks, and also gives a concrete approach under this strategy. The main idea of this strategy is to deploy mesh networks in certain kind of geometric graph, and then design a geometric routing protocol by exploiting the routing properties of this graph. The proposed concrete approach comprises two parts: (1) a topology generation algorithm based on Delaunay triangulations and (2) a geometric routing protocol based on the greedy forwarding algorithm. Both parts are characterized by simplicity and appealing properties, with formal proofs provided when possible. The simulation results validate our proposed approach. [ABSTRACT FROM AUTHOR]

Energy-efficient computing has now become a key challenge not only for data-center operations, but also for many other energy-driven systems, with the focus on reducing of all energy-related costs, and operational expenses, as well as its corresponding and environmental impacts. However, current intelligent data models are typically performance driven. For instance, most data-driven machine-learning approaches are often known to require high computational cost in order to find the global optima. Designing more accurate intelligent data models to satisfy the market needs will hence lead to a higher likelihood of energy waste due to the increased computational cost. This paper thus introduces an energy-efficient framework for large-scale data modeling and classification/prediction. It can achieve a predictive accuracy comparable to or better than the state-of-the-art machine-learning models, while at the same time, maintaining a low computational cost when dealing with large-scale data. The effectiveness of the proposed approaches has been demonstrated by our experiments with two large-scale KDD data sets: Mtv-1 and Mtv-2. [ABSTRACT FROM AUTHOR]

The article announces the appointments of several Associate Editors of the periodical "IEEE Transactions On Computers," including Habib M. Ammari, Krishnendu Chakrabarty and Patrick Schaumont.

Abstract: Time synchronization is a critical component in any wireless sensor network (WSN). In terms of energy consumption, on-demand time synchronization is better than continuous synchronization. However, currently existing on-demand time synchronization protocols have a very low accuracy and very strong spatial accumulative effect. These features are not suitable for several types of WSN applications, such as applications with stringent temporal requirements, or applications that have a large spatial region of interest. In this paper, we propose an on-demand time synchronization protocol, named AOTSP (Accurate On-demand Time Synchronization Protocol), which differs from other protocols of the same category by having the following advantages, as shown in our theoretical analysis and simulation results: (1) weak spatial accumulative effect; (2) fairly low communication cost; (3) low computational complexity; (4) high accuracy; (5) high scalability. Such features make AOTSP a suitable time synchronization protocol for a broad range of WSN applications. [Copyright &y& Elsevier]

The energy consumption of under-utilized resources, particularly in a cloud environment, accounts for a substantial amount of the actual energy use. Inherently, a resource allocation strategy that takes into account resource utilization would lead to a better energy efficiency; this, in clouds, extends further with virtualization technologies in that tasks can be easily consolidated. Task consolidation is an effective method to increase resource utilization and in turn reduces energy consumption. Recent studies identified that server energy consumption scales linearly with (processor) resource utilization. This encouraging fact further highlights the significant contribution of task consolidation to the reduction in energy consumption. However, task consolidation can also lead to the freeing up of resources that can sit idling yet still drawing power. There have been some notable efforts to reduce idle power draw, typically by putting computer resources into some form of sleep/power-saving mode. In this paper, we present two energy-conscious task consolidation heuristics, which aim to maximize resource utilization and explicitly take into account both active and idle energy consumption. Our heuristics assign each task to the resource on which the energy consumption for executing the task is explicitly or implicitly minimized without the performance degradation of that task. Based on our experimental results, our heuristics demonstrate their promising energy-saving capability. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents a new approach, namely Intelligent Fuzzy Online Location Management Strategy (IFOLMS), based on Fuzzy clustering techniques to solve the mobile location management problem. Using a Fuzzy location estimator in this technique, mobile users’ past movements are used in making future paging decisions by the network. IFOLMS has the potential to lead to massive savings in the number of network signal transactions that must be made to locate users. Performance of the proposed approach has been measured by using several test networks; it shows promising results — around 50% reduction in network cost — when compared to many of the existing location management techniques (including GSM). Results also provide new insights into the mobility management problem and its associated performance issues. [Copyright &y& Elsevier]

This work reviews a number of algorithms to solve the location management problem in mobile networks for both static and dynamic scenarios. In the static mode, results of five algorithms are used to highlight the pros and cons for each algorithm. These results provide new insight into the mobility management problem that can influence the design of future wireless networks. In the dynamic mode in which mobile users' past movement patterns are used in making future paging decisions by the network, the performance of an online location management algorithm is examined under different deployment setups. Performance results of this algorithm show its advantages over the currently implemented and/or proposed static location management systems (including GSM). [ABSTRACT FROM AUTHOR]

Performance perturbations are a natural phenomenon in volunteer computing systems. Scheduling parallel applications with precedence-constraints is emerging as a new challenge in these systems. In this paper, we propose two novel robust task scheduling heuristics, which identify best task-resource matches in terms of makespan and robustness. Our approach for both heuristics is based on a proactive reallocation (or schedule expansion) scheme enabling output schedules to tolerate a certain degree of performance degradation. Schedules are initially generated by focusing on their makespan. These schedules are scrutinized for possible rescheduling using additional volunteer computing resources to increase their robustness. Specifically, their robustness is improved by maximizing either the total allowable delay time or the minimum relative allowable delay time over all allocated volunteer resources. Allowable delay times may occur due to precedence constraints. In this paper, two proposed heuristics are evaluated with an extensive set of simulations. Based on simulation results, our approach significantly contributes to improving the robustness of the resulting schedules. [ABSTRACT FROM AUTHOR]

Abstract: Energy usage and its associated costs have taken on a new level of significance in recent years. Globally, energy costs that include the cooling of server rooms are now comparable to hardware costs, and these costs are on the increase with the rising cost of energy. As a result, there are efforts worldwide to design more efficient scheduling algorithms. Such scheduling algorithm for grids is further complicated by the fact that the different sites in a grid system are likely to have different ownerships. As such, it is not enough to simply minimize the total energy usage in the grid; instead one needs to simultaneously minimize energy usage between all the different providers in the grid. Apart from the multitude of ownerships of the different sites, a grid differs from traditional high performance computing systems in the heterogeneity of the computing nodes as well as the communication links that connect the different nodes together. In this paper, we propose a cooperative, power-aware game theoretic solution to the job scheduling problem in grids. We discuss our cooperative game model and present the structure of the Nash Bargaining Solution. Our proposed scheduling scheme maintains a specified Quality of Service (QoS) level and minimizes energy usage between all the providers simultaneously; energy usage is kept at a level that is sufficient to maintain the desired QoS level. Further, the proposed algorithm is fair to all users, and has robust performance against inaccuracies in performance prediction information. [Copyright &y& Elsevier]

This work reviews a number of bio-inspired algorithms and examines their suitability for solving the location management problem by using the paging cell and location area schemes. A comparison is performed using three test networks using realistic user movement patterns. The results provide new insight into the mobility management problem that can influence the design of future wireless networks. [ABSTRACT FROM AUTHOR]

Background: Multiple Sequence Alignment (MSA) has always been an active area of research in Bioinformatics. MSA is mainly focused on discovering biologically meaningful relationships among different sequences or proteins in order to investigate the underlying main characteristics/ functions. This information is also used to generate phylogenetic trees. Results: This paper presents a novel approach, namely RBT-GA, to solve the MSA problem using a hybrid solution methodology combining the Rubber Band Technique (RBT) and the Genetic Algorithm (GA) metaheuristic. RBT is inspired by the behavior of an elastic Rubber Band (RB) on a plate with several poles, which is analogues to locations in the input sequences that could potentially be biologically related. A GA attempts to mimic the evolutionary processes of life in order to locate optimal solutions in an often very complex landscape. RBT-GA is a population based optimization algorithm designed to find the optimal alignment for a set of input protein sequences. In this novel technique, each alignment answer is modeled as a chromosome consisting of several poles in the RBT framework. These poles resemble locations in the input sequences that are most likely to be correlated and/or biologically related. A GA-based optimization process improves these chromosomes gradually yielding a set of mostly optimal answers for the MSA problem. Conclusion: RBT-GA is tested with one of the well-known benchmarks suites (BALiBASE 2.0) in this area. The obtained results show that the superiority of the proposed technique even in the case of formidable sequences. [ABSTRACT FROM AUTHOR]

A grid differs from traditional high-performance computing systems in the heterogeneity of the computing nodes, as well as the communication links that connect the different nodes together. In grids, there exist users and service providers. The service providers provide the service for jobs that the users generate. Typically, the amount of jobs generated by all of the users is more than any single provider can handle alone with any acceptable quality of service (QoS). As such, the service providers need to cooperate and allocate jobs among them so that each is providing an acceptable QoS to their customers. QoS is of particular concern to service providers as it directly affects customers' satisfaction and loyalty. In this paper, we propose a game theoretic solution to the QoS sensitive grid job allocation problem. We model the QoS-based grid job allocation problem as a cooperative game and present the structure of the Nash Bargaining Solution. The proposed algorithm is fair to all users and represents a Pareto optimal solution to the QoS objective. One advantage of our scheme is the relatively low overhead and robust performance against inaccuracies in performance prediction information. [ABSTRACT FROM AUTHOR]

Despite numerous efforts, task scheduling remains one of the most challenging problems in heterogeneous computing environments. In this paper, we propose a new state transition (ST) method, called the Duplication-based State Transition (DST) method, which is specially designed for metaheuristics that can be used for the task-scheduling problem in heterogeneous computing environments. ST in metaheuristics is a key component that takes charge of generating variants of a given state. The DST method produces a new state by first overlapping randomly generated states with the current state, and then, the resultant state is refined by removing ineffectual tasks. The proposed method is incorporated into three different metaheuristics: genetic algorithms (GAs), simulated annealing (SA), and artificial immune systems (AISs). They are experimentally evaluated and are also compared with existing algorithms. The experimental results confirm DST's promising impact on the performance of the metaheuristics. [ABSTRACT FROM AUTHOR]

Load balancing is a very important and complex problem in computational grids. A computational grid differs from traditional high-performance computing systems in the heterogeneity of the computing nodes, as well as the communication links that connect the different nodes together. There is a need to develop algorithms that can capture this complexity yet can be easily implemented and used to solve a wide range of load-balancing scenarios. In this paper, we propose a game-theoretic solution to the grid load-balancing problem. The algorithm developed combines the inherent efficiency of the centralized approach and the fault-tolerant nature of the distributed, decentralized approach. We model the grid load-balancing problem as a noncooperative game, whereby the objective is to reach the Nash equilibrium. Experiments were conducted to show the applicability of the proposed approaches. One advantage of our scheme is the relatively low overhead and robust performance against inaccuracies in performance prediction information. [ABSTRACT FROM AUTHOR]

Abstract: Load balancing is a very important and complex problem in computational grids. A computational grid differs from traditional high performance computing systems in the heterogeneity of the computing nodes and communication links, as well as background workloads that may be present in the computing nodes. There is a need to develop algorithms that could capture this complexity yet can be easily implemented and used to solve a wide range of load balancing scenarios. Artificial life techniques have been used to solve a wide range of complex problems in recent times. The power of these techniques stems from their capability in searching large search spaces, which arise in many combinatorial optimization problems, very efficiently. This paper studies several well-known artificial life techniques to gauge their suitability for solving grid load balancing problems. Due to their popularity and robustness, a genetic algorithm (GA) and tabu search (TS) are used to solve the grid load balancing problem. The effectiveness of each algorithm is shown for a number of test problems, especially when prediction information is not fully accurate. Performance comparisons with Min–min, Max–min, and Sufferage are also discussed. [Copyright &y& Elsevier]

Over the past decade, the grid has emerged as an attractive platform to tackle various large-scale problems, especially in science and engineering. One primary issue associated with the efficient and effective utilization of heterogeneous resources in a grid is scheduling. Grid scheduling involves a number of challenging issues, mainly due to the dynamic nature of the grid. There are only a handful of scheduling schemes for grid environments that realistically deal with this dynamic nature that have been proposed in the literature. In this paper, two novel scheduling algorithms, called the Shared-Input-data-based Listing (SIL) algorithm and the Multiple Queues with Duplication (MQD) algorithm for bag-of-tasks (BoT) applications in grid environments are proposed. The SIL algorithm targets scheduling data-intensive BoT (DBoT) applications, whereas the MOD algorithm deals with scheduling computationally intensive BoT (CBoT) applications. Their common and primary forte is that they make scheduling decisions without fully accurate performance prediction information. Another point to note is that both scheduling algorithms adopt task duplication as an attempt to reduce serious schedule increases. Our evaluation study employs a number of experiments with various simulation settings. The results show the practicability and competitiveness of our algorithms when compared to existing methods. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents a novel approach based on clustering algorithms in combination with the location area (LA) scheme to solve the mobility management problem. Users’ movement history is used by the network to predict future paging decisions. This approach integrates the LAs scheme and efficient clustering algorithms to find a network topology which can lead to massive savings in the number of signals made to locate users in the network. The approach is tested with several networks to show its advantages to the current GSM standards. The results provide new insights into the mobility management problem. [Copyright &y& Elsevier]

Background: Knowledge of protein domain boundaries is critical for the characterisation and understanding of protein function. The ability to identify domains without the knowledge of the structure — by using sequence information only — is an essential step in many types of protein analyses. In this present study, we demonstrate that the performance of DomainDiscovery is improved significantly by including the inter-domain linker index value for domain identification from sequence-based information. Improved DomainDiscovery uses a Support Vector Machine (SVM) approach and a unique training dataset built on the principle of consensus among experts in defining domains in protein structure. The SVM was trained using a PSSM (Position Specific Scoring Matrix), secondary structure, solvent accessibility information and inter-domain linker index to detect possible domain boundaries for a target sequence. Results: Improved DomainDiscovery is compared with other methods by benchmarking against a structurally non-redundant dataset and also CASP5 targets. Improved DomainDiscovery achieves 70% accuracy for domain boundary identification in multi-domains proteins. Conclusion: Improved DomainDiscovery compares favourably to the performance of other methods and excels in the identification of domain boundaries for multi-domain proteins as a result of introducing support vector machine with benchmarḵ2 dataset. [ABSTRACT FROM AUTHOR]

The scheduling problem deals with the optimal assignment of a set of tasks to processing elements in a distributed system such that the total execution time is minimized. One approach for solving the scheduling problem is task clustering. This involves assigning tasks to clusters where each cluster is run on a single processor. This paper aims to show the feasibility of using Genetic Algorithms for task clustering to solve the scheduling problem. Genetic Algorithms are robust optimization and search techniques that are used in this work to solve the task-clustering problem. The proposed approach shows great promise to solve the clustering problem for a wide range of clustering instances. [ABSTRACT FROM AUTHOR]

Abstract: Network behavior is the set of observations or measurements that can be made about a network over time. The growth of network-based computing and the Internet have ensured that networks can no longer be considered in isolation, as events external to a particular network increasingly impact its behavior. Network management requires that information be known about these events, a task that is not always possible. We present a modeling strategy that takes partial information about a network and uses it to predict the behavior in unmonitored areas. This implementation is based on a meta-heuristic (genetic algorithm), and uses IP-packet information as well as a limited understanding of the external topology. This is then used to model the full topology, routing tables and traffic for the entire network at periodic intervals. The system was tested using the ns-2 network simulator and a Java implementation on a series of cases. The results showed a reasonable level of accuracy in predicting traffic and topology. Performance increased under system load, and at no point did the system generate any additional network traffic. This provides an efficient and effective strategy for network management. [Copyright &y& Elsevier]

Voronoi diagram is one of the most fundamental and important geometric data structures. Voronoi diagram was historically defined for a set of points on the plane. The diagram partitions the plane into regions, one per site. The region of a site s consists of all points closer to s than to any other sites on the plane. Concepts of Voronoi diagram are often attributed to Voronoi (J. Reine Angew. Math. 133 (1907) 97) and Dirichlet (J. Reine Angew. Math. 40 (1850) 209). As a result of these early works, often the name Voronoi diagram and Dirichlet tessellation is used. Due to the importance of Voronoi diagrams, it is important that algorithms are devised to compute these structures in an efficient manner. Of course, this will create new opportunities for the applicability of these data structures. Towards this end, this paper presents new results for the computation of Voronoi diagrams for a set of n points, or n disjoint circles on the plane, on a mesh with multiple broadcasting (MMB) of size n×n. The algorithm runs in O(log2 n) time. [Copyright &y& Elsevier]

The channel allocation problem (CAP) that involves the allocation a disjoint set of channels to meet the call demands for a cellular network is an NP-complete combinatorial optimisation problem [1]. The CAP can be viewed as: static (during the initial design/planning of the cellular network) and dynamic (when the network is operational). This paper presents a new algorithm designed to solve the online call control problem. This algorithm is a modified version of the maximum channel packing channel allocation (MCPCA) scheme, proposed by [19]. The original MCPCA scheme aims at maximising the reuse of channels (i.e. dense packing) and simulation results showed that it is more efficient than fixed channel assignment (FCA) or borrowing channel assignment (BCA) schemes for solving the class of minimum blocking frequency assignment problem (MB-FAP) [2]. The new algorithm, entitled maximum channel packing channel assignment with re-assignment (MCPCA-RA), takes the dense packing mechanism further by allowing a neighbouring cell to re-assigns a channel to an existing call and releases the channel previously used to the new request. Simulation results that the number of blocked calls is reduced by an average of 6% compared to the original MCPCA scheme, but at the cost of extra computations due to the reassignment mechanism. Since the reassignment computations only involve cells in the local neighbourhood and can be compute in a parallel manner, MCPCA-RA algorithm is practical and efficient in real-time. [ABSTRACT FROM AUTHOR]

Location management is a very important and complex problem in today's mobile computing environments. There is a need to develop algorithms that could capture this complexity yet can be easily implemented and used to solve a wide range of location management scenarios. Artificial life techniques have been used to solve a wide range of complex problems in recent times. The power of these techniques stems from their capability in searching large search spaces, which arise in many combinatorial optimization problems, very efficiently. This paper compares several well-known artificial life techniques to gauge their suitability for solving location management problems. Due to their popularity and robustness, a Genetic algorithm (GA), tabu search (TS), and ant colony algorithm (ACA) are used to solve the reporting cells planning problem. In the reporting cell location management scheme, some cells in the network are designated as reporting cells; mobile terminals update their positions (location update) upon entering one of these reporting cells. To create such a planner, a GA, TS, as well as several different AC algorithms are implemented. The effectiveness of each algorithm is shown for a number of test problems. [ABSTRACT FROM AUTHOR]

The last thirty years have seen tremendous growth in research in mobile telecommunications. However, interest in mobile computing, which includes mobile telephony and more, has increased over the last ten years. Nevertheless, most of the research on mobile computing addresses the "engineering" issues and the electronic componentary required for building mobile systems. On the other hand, algorithmics research in mobile computing is still in its infancy, and only dates back to few years. The elegance and terseness that exist today in algorithmics, especially parallel computing algorirthmics, can be brought to bear on research on "mobile computing algorithmics". However, mobility brings to the stage a whole gamut of new problems that needs to be addressed in order to develop new and efficient parallel algorithms that can be used to solve mobile computing problems. [ABSTRACT FROM AUTHOR]

With the increase in global wireless communication, there is a need for efficient network management strategies. Location management in a mobile network involves keeping track of mobile host (MH) cell locations. MHs perform location updates to inform the network of their location. When a call arrives for an MH, the network uses its last known cell location and a paging strategy to find that host. Current location management techniques do not consider host-mobility patterns or call arrival rates. This paper describes a selective update strategy that is modeled on the characteristics of a network, such as, topology, host mobility patterns and connection request rates. Then, a genetic algorithm is used to solve the location management problem that involves the search of a large solution space. The aim of the work is to determine whether genetic algorithms can be applied successfully to solve this problem, and to evaluate their efficiency in solving this class of optimization problems. Results from the selective update strategy show improvements over alternative algorithms. The location management optimization problem is shown to be well adapted to the workings of the genetic algorithm. The proposed solution also saves power, processing time, and network bandwidth. [Copyright &y& Elsevier]

Location management is a very important and complex problem in mobile computing. There is a need to develop algorithms that could capture this complexity yet can be easily implemented and used to solve a wide range of location management scenarios. This paper investigates the use of cellular automata (CA) combined with genetic algorithms to create an evolving parallel reporting cells planning algorithm. In the reporting cell location management scheme, some cells in the network are designated as reporting cells; mobile terminals update their positions (location update) upon entering one of these reporting cells. To create such an evolving CA system, cells in the network are mapped to cellular units of the CA and neighborhoods for the CA is selected. GA is then used to discover efficient CA transition rules. The effectiveness of the GA and of the discovered CA rules is shown for a number of test problems. [ABSTRACT FROM AUTHOR]

Presents a novel approach in designing a cellular automata-based multiprocessor scheduling algorithms. Concept of the automata-based scheduler; Allocation of tasks in the system; Normal operating mode of the scheduler.

This paper highlights the potential of using genetic algorithms to solve cellular resource allocation problems. Generally, cellular resource allocation deals with how and when to allocate radio frequency channels to mobile hosts. The objective in this work is to gauge how well a GA-based channel borrower performs when compared to a greedy borrowing heuristic. This is needed to establish how suited GA-like (stochastic search) algorithms are for the solution of optimization problems in mobile computing environments. This involves the creation of a simple mobile networking resource environment and design of a GA-based channel borrower that works within this environment. A simulation environment is also built to compare the performance of the GA-based channel-borrowing method with the heuristic. To enhance the performance of the GA, extra attention is paid to developing an improved mutation operator. The performance of the new operator is evaluated against the heuristic borrowing scheme. For a real-time implementation, the GA needs to have the properties of a micro GA strategy [4]. This involves making improvements to the crossover operator and evaluation procedure so the GA can converge to a "good" solution rapidly. [ABSTRACT FROM AUTHOR]

Recursive least squares (RLS) is a popular iterative method used for the modeling of systems while in operation. RLS provides an estimate for unknown parameters of a system based on some known parameters and inputs and outputs of that system. This technique is used frequently in digital signal processing and control applications, where it is not possible to completely determine the current state of the system. The RLS procedure incurs intensive computations in every iteration of the algorithm. To implement RLS in situ at a reasonable sampling rate, the complexity of the system''s model must be reduced, or the available computing power must be increased. This paper examines methods for increasing the computing power by implementing RLS algorithms on a parallel processing platform. While there has been a large body of research on using parallel processors for the computation of adaptive algorithms, little of this research has examined fault tolerant aspects. As fault tolerance is a critical aspect of any real-time system, this work will examine some factors that should be considered when implementing a real-time adaptive algorithm on a parallel processor system. [Copyright &y& Elsevier]

Elaborates on the application of the concept of task scheduling in computer science. Advantages of dynamic scheduling; Scheduling scenarios for parallel computing systems.

Examines the use of genetic algorithms for dynamic load-balancing. Role of load-balancing operations in the parallel and distributed computing systems; Employment of heuristic methods in load-balancing researches; Effects of load-balancing issues on the success of genetic-based load balancing algorithm.

Dependency-aware jobs, such as the big data analytic workflows, are commonly executed on the cloud. They are compiled to directed acyclic graphs, with tasks linked in regarding the dependency. The cloud scheduler, which maintains a large number of resources, is responsible to execute tasks in parallel. To resolve the complex dependencies, Deep Reinforcement Learning (DRL) based schedulers are widely applied. However, we find that the DRL-based schedulers are vulnerable to the perturbations in the input jobs and may generate falsified decisions to benefit a particular job while delaying the others. By perturbation, we mean a slight adjustment to the job's node features or dependencies, while not changing its functionality. In this paper, we first explore the vulnerability of DRL-based schedulers to job perturbations without accessing the information of the DRL models used in the scheduler. We devise the black-box perturbation system, in which, a proxy model is trained to mimic the DRL-based scheduling policy. We show that the high-faith proxy model can help to craft effective perturbations. The DRL-based schedulers can be as high as 60% likely to be badly affected by the perturbations. Then, we investigate the solution to improve the robustness of DRL-based schedulers to such perturbations. We propose an adversarial training framework to force the neural model to adapt to the perturbation patterns during training so as to eliminate the potential damage during applications. Experiments show that the adversarial-trained scheduler is more robust, reducing the chance of being affected to 3-fold less and the potential bad effects halved. [ABSTRACT FROM AUTHOR]

Presents an alternative paradigm based on genetic algorithms to efficiently solve the scheduling problem without the need to apply any restricted assumptions that are problem-specific. Comparison between the proposed algorithm and well-known list scheduling heuristics; Conditions under which the genetic algorithm performs best; Parallel processing.

Focuses on the functioning of parallel processor systems through reinforcement-based scheduling (RBS). Information on supervised versus reinforcement learning; Detailed information on reinforcement learning; Indepth look at the algorithm for the RBS; More detailed information on the parallel processor systems and RBS.

Investigates the interlocking properties between sets of `IOTA' in various applications. Reference to some of the applications of the IOTA interlocking systems; Information on how the system works.

Robotic tasks (e.g., kinematics and dynamics) are computationally expensive. The majority of these tasks must be computed in real-time to meet the high sampling rate modes of operations. Recently parallel processing has been used to speed up these computations. In this work, we propose a graph-based algorithm to map computational tasks onto multiple instruction--multiple data (MIMD) type of architectures. The algorithm automatically generates the task graph of a given task. Then an annealing procedure is used to allocate the generated subtasks to different processors, taking into account the network topology and the communication constraints. Moreover, the proposed technique is simple, flexible, and computationally viable. The efficiency of the algorithm is demonstrated by a case study with good results. [ABSTRACT FROM AUTHOR]

Focuses on the efficient parallel implementation of systems of numerically intensive nature over loosely coupled multiprocessor architectures. Formal mathematical definition of the problem concern; Main two components of the engine; Different modules of the task graph generator; Schedules of two different computational models on parallel architectures; Conclusions.

This installment of Computer's series highlighting the work published in IEEE Computer Society journals comes from IEEE Transactions on Sustainable Computing. [ABSTRACT FROM AUTHOR]

Interest in processing big data has increased rapidly to gain insights that can transform businesses, government policies, and research outcomes. This has led to advancement in communication, programming, and processing technologies, including cloud computing services and technologies such as Hadoop, Spark, and Storm. This trend also affects the needs of analytical applications, which are no longer monolithic but composed of several individual analytical steps running in the form of a workflow. These big data workflows are vastly different in nature from traditional workflows. Researchers are currently facing the challenge of how to orchestrate and manage the execution of such workflows. In this article, we discuss in detail orchestration requirements of these workflows as well as the challenges in achieving these requirements. We also survey current trends and research that supports orchestration of big data workflows and identify open research challenges to guide future developments in this area. [ABSTRACT FROM AUTHOR]

The pervasive use of virtualization techniques in today's datacenters poses challenges in power monitoring since it is not possible to directly measure the power consumption of a virtual entity such as a virtual machine (VM) and a container. In this paper, we present cWatts++, a lightweight virtual power meter that enables accurate power usage measurement in virtualized computing environments such as VMs and containers of Cloud data centers. At the core of cWatts++ is its application-agnostic power model. To this end, we devise two power models (eventModel and raplModel) that are driven by CPU event counters and the Running Average Power Limit (RAPL) feature of modern Intel CPUs, respectively. While eventModel is more generic and, thus, applicable to a wide range of workloads, raplModel is particularly good for CPU-bound workloads. We have evaluated cWatts++ with its two power models in a real system using the PARSEC benchmark suite and our in-house benchmarks. Our evaluation study demonstrates that these power models have an average error of 4.55 and 1.25 percent, respectively, compared with actual power usage measurements of a real power meter, Cabac Power-Mate. [ABSTRACT FROM AUTHOR]

This installment of Computer's series highlighting the work published in IEEE Computer Society journals comes from IEEE Transactions on Computers. [ABSTRACT FROM PUBLISHER]

Timely discovering and acquiring information from incremental data on the Internet is a hot topic in a big data era. This paper presents a distributed incremental information acquisition model for large-scale text data. To obtain a lower false positive rate and higher efficiency of the traditional Bloom filter, a distributed multidimensional Bloom filter is designed and proposed to cope with the deduplication of large-scale Web URL text data. Three methods related to Bloom filter were compared based on the false positive rate and response efficiency. The results show that the distributed incremental information acquisition model for large-scale text data can achieve a high duplicate removal rate with a lower false positive rate. [ABSTRACT FROM AUTHOR]