Bag Of Tasks Research Articles

A delay-based dynamic scheduling algorithm for bag-of-task workflows with stochastic task execution times in clouds

Bag-of-Tasks (BoT) workflows are widespread in many big data analysis fields. However, there are very few cloud resource provisioning and scheduling algorithms tailored for BoT workflows. Furthermore, existing algorithms fail to consider the stochastic task execution times of BoT workflows which leads to deadline violations and increased resource renting costs. In this paper, we propose a dynamic cloud resource provisioning and scheduling algorithm which aims to fulfill the workflow deadline by using the sum of task execution time expectation and standard deviation to estimate real task execution times. A bag-based delay scheduling strategy and a single-type based virtual machine interval renting method are presented to decrease the resource renting cost. The proposed algorithm is evaluated using a cloud simulator ElasticSim which is extended from CloudSim. The results show that the dynamic algorithm decreases the resource renting cost while guaranteeing the workflow deadline compared to the existing algorithms.

MLBox: Machine learning box for asymptotic scheduling

As the usability of Cloud-based solutions has increased for various types of users with different needs, from scientists that want to process big data sets collected from sensors or business analysts that want to take decisions based on the huge amount of gathered data to simple users that store or share documents via a Cloud platform, the generated data is increasing more and more. For example, the ATLAS and other detectors at CERN generate petabytes of data and Facebook stores data with a rate of around 600 TB daily. In the current context, efficient scheduling for Big Data applications is a challenge and an appropriate scheduling technique is required for different types of incoming requests. In this paper we propose a scheduling algorithm for different types of computation requests: independent tasks, like bag of tasks (BoT) model or tasks with dependencies modeled as directed acyclic graphs (DAG), and they will be scheduled for execution in a Cloud datacenter. The tasks in the requests are scheduled on the available resources using the suitable scheduling algorithm for each request. We rely on a machine learning toolbox, named as MLBox, to find what algorithm should be used for a certain request. We implemented four heuristics for scheduling BoTs and four heuristics for DAGs scheduling and generated the training data for the machine learning algorithm by running multiple traditional scheduling algorithms and selecting the ‘best’ one for a given request. We evaluate the performance by comparing the scheduling of different tasks requests using some of the traditional algorithms and our machine learning based scheduling algorithm.

Simple, efficient allocation of modelling runs on heterogeneous clusters with MPI

In scientific modelling and computation, the choice of an appropriate method for allocating tasks for parallel processing depends on the computational setting and on the nature of the computation. The allocation of independent but similar computational tasks, such as modelling runs or Monte Carlo trials, among the nodes of a heterogeneous computational cluster is a special case that has not been specifically evaluated previously. A simulation study shows that a method of on-demand (that is, worker-initiated) pulling from a bag of tasks in this case leads to reliably short makespans for computational jobs despite heterogeneity both within and between cluster nodes. A simple reference implementation in the C programming language with the Message Passing Interface (MPI) is provided.

A shared cache-aware Task scheduling strategy for multi-core systems

In the past few years, multi-core processors incorporating four, six, eight, or more cores on a single die have become ubiquitous. Those cores, having their own private caches, often share a higher level cache memory, which leads to compete among different tasks. This can seriously affect the avera ge performance of multi-core systems as the probability of cache hit could be lowered. In realizing this, we study the problem of scheduling bag-of-tasks (BoT) applications with shared cache constraint on multi-core systems. We first use cache space isolation techniques to divide shared caches into partitions. Then, we give a motivational example and outline the shared cache aware scheduling problem of multi-core systems. Finally, to provide an optimum solution for this problem, we propose a heuristic shared cache contention aware scheduling (SCAS) algorithm on multi-core systems. Our extensive simulation performance evaluation study clearly demonstrate that our proposed SCAS algorithm outperforms the existing traditional scheduling algorithm Min-min and the modified algorithm MSCAS in terms of schedule length and average response time.

Power-aware Bag-of-Tasks scheduling on heterogeneous platforms

Energy preservation is very important nowadays. A large number of applications in science, engineering, astronomy and business analytics are classified as Bag-of-Tasks (BoT) applications. A BoT is a collection of independent tasks that do not communicate with each other during execution. BoT scheduling has been severely studied from a performance point of view. In this paper, we address the problem of energy-efficient BoT scheduling in a heterogeneous environment with the twin objectives of minimizing finish time and energy consumption. Specifically, we extend two performance-oriented scheduling policies, Min---Min and Max---Min, and propose power-aware centralized scheduling policies that incorporate a dynamic voltage/frequency scaling mechanism and can power on and off unneeded computing nodes of a heterogeneous cluster environment using dynamic power management. Additionally, to evaluate the system using a more realistic workload, high-priority tasks with and without time-constraints are also submitted. A series of simulation experiments conducted, show that we can achieve significant energy savings without affecting significantly the execution of BoTs and high-priority tasks. Additional experiments on a real system also confirmed the effectiveness of our policies.

Approaches for Efficient Workflow Planning and Execution of Scientific Application with Cloud Resources to Maximize Throughput within a Limited Time Limit

In Cloud Environment, Resources available to the client on demand with pay per usage. Higher throughput with minimal execution time can reduce the budget cost for client as well as can never violate Service Level Agreement (SLA). For scientific application like weather forecasting maximum calculation within a limited time scope must be requirement. Prediction of incoming future work and its parallel execution with unlimited cloud resources can fulfil our requirement. Various techniques should be developed for this strategy. Usage Pattern [1] scheme use knowledge base for predicting future jobs by pattern matching which help to reduce start-up overhead. Bag of Task [2] scheme provide concept of simultaneous execution of scientific tasks with variable resources on available cloud. Resources can be increased or decreased on demand which can reduce the overall cloud rent cost. Our approach suggests a technique to combine usage pattern with bag of task which can provide efficient result with higher throughput in minimum time. Predict the future work, estimate execution time, divide jobs in small tasks and execute them parallel with on-demand variable resources can provide good results.

Energy-Efficient Stochastic Task Scheduling on Heterogeneous Computing Systems

In the past few years, with the rapid development of heterogeneous computing systems (HCS), the issue of energy consumption has attracted a great deal of attention. How to reduce energy consumption is currently a critical issue in designing HCS. In response to this challenge, many energy-aware scheduling algorithms have been developed primarily using the dynamic voltage-frequency scaling (DVFS) capability which has been incorporated into recent commodity processors. However, these techniques are unsatisfactory in minimizing both schedule length and energy consumption. Furthermore, most algorithms schedule tasks according to their average-case execution times and do not consider task execution times with probability distributions in the real-world. In realizing this, we study the problem of scheduling a bag-of-tasks (BoT) application, made of a collection of independent stochastic tasks with normal distributions of task execution times, on a heterogeneous platform with deadline and energy consumption budget constraints. We build execution time and energy consumption models for stochastic tasks on a single processor. We derive the expected value and variance of schedule length on HCS by Clark's equations. We formulate our stochastic task scheduling problem as a linear programming problem, in which we maximize the weighted probability of combined schedule length and energy consumption metric under deadline and energy consumption budget constraints. We propose a heuristic energy-aware stochastic task scheduling algorithm called ESTS to solve this problem. Our algorithm can achieve high scheduling performance for BoT applications with low time complexity $O(n(M+\log n))$ , where $n$ is the number of tasks and $M$ is the total number of processor frequencies. Our extensive simulations for performance evaluation based on randomly generated stochastic applications and real-world applications clearly demonstrate that our proposed heuristic algorithm can improve the weighted probability that both the deadline and the energy consumption budget constraints can be met, and has the capability of balancing between schedule length and energy consumption.

Joint Dimensioning of Server and Network Infrastructure for Resilient Optical Grids/Clouds

We address the dimensioning of infrastructure, comprising both network and server resources, for large-scale decentralized distributed systems such as grids or clouds. We design the resulting grid/cloud to be resilient against network link or server failures. To this end, we exploit relocation: Under failure conditions, a grid job or cloud virtual machine may be served at an alternate destination (i.e., different from the one under failure-free conditions). We thus consider grid/cloud requests to have a known origin, but assume a degree of freedom as to where they end up being served, which is the case for grid applications of the bag-of-tasks (BoT) type or hosted virtual machines in the cloud case. We present a generic methodology based on integer linear programming (ILP) that: chooses a given number of sites in a given network topology where to install server infrastructure; and determines the amount of both network and server capacity to cater for both the failure-free scenario and failures of links or nodes. For the latter, we consider either failure-independent (FID) or failure-dependent (FD) recovery. Case studies on European-scale networks show that relocation allows considerable reduction of the total amount of network and server resources, especially in sparse topologies and for higher numbers of server sites. Adopting a failure-dependent backup routing strategy does lead to lower resource dimensions, but only when we adopt relocation (especially for a high number of server sites): Without exploiting relocation, potential savings of FD versus FID are not meaningful.

Pareto-Optimal Cloud Bursting

Large-scale Bag-of-Tasks (BoT) applications are characterized by their massively parallel, yet independent operations. The use of resources in public clouds to dynamically expand the capacity of a private computer system might be an appealing alternative to cope with such massive parallelism. To fully realize the benefit of this `cloud bursting', the performance to cost ratio (or cost efficiency) must be thoroughly studied and incorporated into scheduling and resource allocation strategies. In this paper, we present PANDA, a framework for static scheduling BoT applications across resources in both private and public clouds. The framework at the core incorporates a fully polynomial-time approximation scheme (FPTAS) as a novel scheduling algorithm, which generates schedules with the best trade-off point between cost and performance; hence Pareto-optimality. We have theoretically discussed the complexity and correctness of our algorithms, and experimentally verified their efficacy and practicality using ISOMAP-a widely-used nonlinear manifold method as a real-world BoT application. Our evaluation conducted in a 'multi-cloud' environment of our 40-core private system and Amazon EC2 public cloud demonstrates the scheduling quality of PANDA is guaranteed to be within a measurable distance from the optimal solution. Results obtained from our experiments show such quality is 8 percent or less from the optimum. We also show the sensitivity and robustness of our scheduling solutions against performance errors in both resources and applications.

Cloud Computing Networks with Poisson Arrival ProcessDynamic Resource Allocation

In this article we considered a cloud computing network model which acts as a tool for dynamic resource allocation problem in cloud computing. Every successful servicing of cloud paradigm need an optimal resource allocation in turn to handle the problem of Bag of tasks (BoT). We assume that the allocation of BoT's are done by two stages of process namely classification according to service level agreement (SLA) of BoT and service providing (SaaS, PaaS and IaaS). After classification the request is routed to any one of the service provider with corresponding probability, Thus the Cloud Computing Networks (CCN) become a general open Jackson Queuing Network system performance measures are obtained to study the efficiency of the CCN.

Requirement-Aware Scheduling of Bag-of-Tasks Applications on Grids with Dynamic Resilience

Grids have been extensively deployed to handle various scientific and engineering applications that can be structured as bag-of-tasks (BoT). The scheduling of BoT applications on Grids is an important issue for achieving high performance. Grid scheduling involves a number of challenging issues, mainly due to the dynamic nature of the Grid. To deal with this dynamic nature, in this paper, we propose an online scheduling algorithm called prudent algorithm with replication (PAR) for scheduling Grid applications. PAR is shown to prudently make scheduling decisions in such a way that it can tolerate inaccurate performance predictions. Another point to note is that PAR adopts task duplication as an attempt to reduce serious schedule increases. Moreover, since the applications to be performed may widely vary in terms of their required hardware and software, we also capture the loads' various processing requirements in our algorithms, a unique feature that is applicable for running proprietary applications only on certain eligible processing nodes. Thus, in our problem formulation each application can only be processed by certain processors as both the applications and processing nodes are heterogeneous. We then present a task selection policy, referred to as requirement-aware load selection (RALS) policy to handle the contention of multiple applications that have various processing requirements but share the same computing resources. Based on RALS and PAR, we develop two scheduling algorithms: requirement-aware prudent algorithm with replication (RAPAR), and requirement-aware knowledge-free algorithm with replication (RAKAR). RAPAR and RAKAR address the scheduling of multiple BoT applications with heterogeneous processing requirements on Grids. RAPAR works in scenarios where inaccurate performance prediction information is provided whereas RAKAR works without any prediction information. Performance evaluation results are presented to demonstrate the effectiveness and competitiveness of our approaches when compared to existing algorithms.

Fair scheduling of bag-of-tasks applications using distributed Lagrangian optimization

Large scale distributed systems typically comprise hundreds to millions of entities (applications, users, companies, universities) that have only a partial view of resources (computers, communication links). How to fairly and efficiently share such resources between entities in a distributed way has thus become a critical question.Although not all applications are suitable for execution on large scale distributed computing platform, ideal are the Bag-of-Tasks (BoT) applications. Hence a large fraction of jobs in workloads imposed on Grids is made of sequential applications submitted in the form of BoTs. Up until now, mainly simple mechanisms have been used to ensure a fair sharing of resources among these applications. Although these mechanisms are proved to be efficient for CPU-bound applications, they are known to be ineffective in the presence of network-bound applications.A possible answer resorts to Lagrangian optimization and distributed gradient descent. Under certain conditions, the resource sharing problem can be formulated as a global optimization problem, which can be solved by a distributed self-stabilizing supply and demand algorithm. In the last decade, this technique has been applied to design various network protocols (variants of TCP, multi-path network protocols, wireless network protocols) and even distributed algorithms for smart grids.In this article, we explain how to use this technique for fairly scheduling concurrent BoT applications with arbitrary communication-to-computation ratio on a Grid. Yet, application heterogeneity raises severe convergence and stability issues that did not appear in the previous contexts and need to be addressed by non-trivial modifications. The effectiveness of our proposal is assessed through an extensive set of complex and realistic simulations.

SpeQuloS: a QoS service for hybrid and elastic computing infrastructures

The large choice of Distributed Computing Infrastructures (DCIs) available allows users to select and combine their preferred architectures amongst Clusters, Grids, Clouds, Desktop Grids and more. In these hybrid DCIs, elasticity is emerging as a key property. In elastic infrastructures, resources available to execute application continuously vary, either because of application requirements or because of constraints on the infrastructure, such as node volatility. In the former case, there is no guarantee that the computing resources will remain available during the entire execution of an application. In this paper, we show that Bag-of-Tasks (BoT) execution on these "Best-Effort" infrastructures suffer from a drop of the task completion rate at the end of the execution. The SpeQuloS service presented in this paper improves the Quality of Service (QoS) of BoT applications executed on hybrid and elastic infrastructures. SpeQuloS monitors the execution of the BoT, and dynamically supplies fast and reliable Cloud resources when the critical part of the BoT is executed. SpeQuloS offers several features to hybrid DCIs users, such as estimating completion time and execution speedup. Performance evaluation shows that BoT executions can be accelerated by a factor 2, while offloading less than 2.5 % of the workload to the Cloud. We report on several scenarios where SpeQuloS is deployed on hybrid infrastructures featuring a large variety of infrastructures combinations. In the context of the European Desktop Grid Initiative (EDGI), SpeQuloS is operated to improve QoS of Desktop Grids using resources from private Clouds. We present a use case where SpeQuloS uses both EC2 regular and spot instances to decrease the cost of computation while preserving a similar QoS level. Finally, in the last scenario SpeQuloS allows to optimize Grid5000 resources utilization.

Analyzing the impact of elasticity on the profit of cloud computing providers

Bag-of-tasks (BoT) is an important class of scientific applications. These applications are typically comprised of a very large number of tasks that can be executed in parallel in an independent way. Due to its cost associativity property, a public cloud computing environment is, theoretically, the ideal platform to execute BoT applications, since it could allow them to be executed as fast as possible, yet without implying any extra costs for the rapid turnaround achieved. Unfortunately, current public cloud computing providers impose strict limits on the amount of resources that a single user can simultaneously acquire, substantially increasing the response time of large BoT applications. In this paper we analyze the reasons why traditional providers need to impose such a limit. We show that increases in the limit imposed have a severe impact on the profit achieved by providers. This leads to the conclusion that new approaches to deploy cloud computing services are required to properly serve BoT applications.

Large-scale volunteer computing over the Internet

Abstract Cycle sharing over the Internet has increased in popularity during the last decade, with increasingly powerful machines being made available to existing projects. In this paper, we present GiGi-MR, a framework that allows non-expert users to run CPU-intensive jobs on top of volunteer resources over the Internet. GiGi-MR has several distinctive features: it allows non-expert users to easily partition their jobs in several parallel tasks; such Bag-of-Tasks (BoT) are executed in parallel as a set of MapReduce applications; the volunteer resources that provide the best match for the tasks being executed are chosen (using attenuated bloom filters); it provides a portable checkpointing fault-tolerance mechanism based on virtualization; it does not rely exclusively on a central server (or servers) at all times (thus minimizing the bottleneck effect); it deals with malicious participants (possibly byzantine) using an efficient partial replication mechanism to validate the results obtained; and it is compatible with BOINC (one of the most popular open-source software platforms for computing using volunteered resources). We describe GiGi-MR’s architecture and evaluate its performance by executing several MapReduce applications on a wide area testbed. Furthermore, we use micro-benchmarks to assess each one of GiGi-MR’s components independently. The system’s overhead is minimal. When compared to an unmodified volunteer computing system, GiGi-MR obtains a performance increase of over 60 % in application turnaround time, while reducing the bandwidth used by an order of magnitude.

Explicit Allocation Strategy with Deadline and Budget Constraint Algorithm in Bag of Tasks Grid

Problem statement: This study is for effective scheduling of grid job s based on economy for space shared resources in Bag of tasks grid. Gr id Computing aims in combining the power of heterogeneous, geographically distributed, multi-do main computational resources to provide high performance or high throughput. Approach: Space shared resources are parallel supercomputers and clusters of workstations that provides a great amou nt of computational power. These resources require jobs to be specified formally in terms of the amoun t of time (tr) and number of processors (p) needed for execution. Bag-of-Tasks (BoT) is an application consists of several uniprocessor and independent tasks that have no inter-task communications or tas k-dependencies. BoT is highly suitable for executio n in grids. It is capable of tolerating network delay s or faults and does not require formal job submiss ion. The Explicit allocation strategy assigns the formal job parameters (p, tr) to the job requests, minimi zing the overhead on the grid users to provide a formal job specification. This strategy uses adaptive heur istics to determine the parameters based on certain heuris tics, in order to improve throughput. In the propos ed system, explicit allocation strategy combined with Deadline and Budget Constraint (DBC) Cost Time optimization algorithm performs effective schedulin g of the jobs based on the user's quality of servic e (QoS) requirements such as deadline, budget and optimization strategy. Results: The cost-time optimization scheduling allocates the cheapest reso urces to ensure that the deadline can be met and computation is minimized. In case if there are two resources with the same cost, scheduling is done in any affordable resource so that the job gets execut ed as early as possible. Conclusion: The performance of this scheme against the existing system is evalu ated using cost factor (C factor ) and speed up ratio (T speedup ) and this scheme is more effective than the existi ng system.

Coordinated rescheduling of Bag‐of‐Tasks for executions on multiple resource providers

SUMMARYMetaschedulers can distribute parts of a Bag‐of‐Tasks (BoT) application among various resource providers in order to speed up its execution. The expected completion time of the user application is then calculated based on the run‐time estimates of all applications running and waiting for resources. However, because of inaccurate run time estimates, initial schedules are not those that provide users with the earliest completion time. These estimates increase the time distance between the first and last tasks of a BoT application, which increases average user response time, especially in multi‐provider environments. This paper proposes a coordinated rescheduling algorithm to handle inaccurate run‐time estimates when executing BoT applications in multi‐provider environments. The coordinated rescheduling defines which tasks can have start time updated based on the expected completion time of the entire BoT application. We have also evaluated the impact of system‐generated run‐time estimates to schedule BoT applications on multiple providers. We performed experiments using simulations and a real distributed platform, Grid'5000. From our experiments, we obtained reductions of up to 5 and 10% for response time and slowdown metrics, respectively, by using coordinated rescheduling over a traditional rescheduling solution. Moreover, coordinated rescheduling requires little modification of existing scheduling systems. System‐generated predictions, on the other hand, are more complex to be deployed and may not reduce response times as much as coordinated rescheduling. Copyright © 2011 John Wiley & Sons, Ltd.

A2HA—automatic and adaptive host allocation in utility computing for bag-of-tasks

AbstractThere are increasingly more computing problems requiring lengthy parallel computations. For those without access to current cluster or grid infrastructures, a recent and proven viable solution can be found with on-demand utility computing infrastructures, such as Amazon Elastic Compute Cloud (EC2). A relevant class of such problems, Bag-of-Tasks (BoT), can be easily deployed over such infrastructures (to run on pools of virtual computers), if provided with suitable software for host allocation. BoT problems are found in several and relevant scenarios such as image rendering and software testing.In BoT jobs, tasks are mostly independent; thus, they can run in parallel with no communication among them. The number of allocated hosts is relevant as it impacts both the speedup and the cost: if too many hosts are used, the speedup is high but this may not be cost-effective; if too few are used, the cost is low but speedup falls below expectations. For each BoT job, given that there is no prior knowledge of neither the total job processing time nor the time each task takes to complete, it is hard to determine the number of hosts to allocate. Current solutions (e.g., bin-packing algorithms) are not adequate as they require knowing in advance either the time that the next task will take to execute or, for higher efficiency, the time taken by each one of the tasks in each job considered.Thus, we present an algorithm and heuristics that adaptively predicts the number of hosts to be allocated, so that the maximum speedup can be obtained while respecting a given predefined budget. The algorithm and heuristics were simulated against real and theoretical workloads. With the proposed solution, it is possible to obtain speedups in line with the number of allocated hosts, while being charged less than the predefined budget.

Grid Computing Workloads

In the mid 1990s, the grid computing community promised the "compute power grid," a utility computing infrastructure for scientists and engineers. Since then, a variety of grids have been built worldwide, for academic purposes, specific application domains, and general production work. Understanding grid workloads is important for the design and tuning of future grid resource managers and applications, especially in the recent wake of commercial grids and clouds. This article presents an overview of the most important characteristics of grid workloads in the past seven years (2003-2010). Although grid user populations range from tens to hundreds of individuals, a few users dominate each grid's workload both in terms of consumed resources and the number of jobs submitted to the system. Real grid workloads include very few parallel jobs but many independent single-machine jobs (tasks) grouped into single "bags of tasks."

Scalability limits of Bag-of-Tasks applications running on hierarchical platforms

Bag-of-Tasks applications are parallel applications composed of independent (i.e., embarrassingly parallel) tasks, which do not communicate with each other, may depend upon one or more input files, and can be executed in any order. Each file may be input for more than one task. Examples of Bag-of-Tasks (BoT) applications include Monte Carlo simulations, massive searches (such as key breaking), image manipulation applications and data mining algorithms. A common framework to execute BoT applications is the master–slave topology, in which the user machine is used to control the execution of tasks. In this scenario, a large number of concurrent tasks competing for resources (e.g., CPU and communication links) severely limits application execution scalability. This paper is devoted to study the scalability of BoT applications running on multi-node systems (such as clusters and multi-clusters) organized as hierarchical platforms, considering several communication paradigms. Our study employs a set of experiments that involves the simulation of various large-scale platforms. The results presented provide important guidelines for improving the scalability of practical applications.