Elina Pacini

Software Survey: Distributed job scheduling based on Swarm Intelligence: A survey

Scientists and engineers need computational power to satisfy the increasing resource intensive nature of their simulations. For example, running Parameter Sweep Experiments (PSE) involve processing many independent jobs, given by multiple initial configurations (input parameter values) against the same program code. Hence, paradigms like Grid Computing and Cloud Computing are employed for gaining scalability. However, job scheduling in Grid and Cloud environments represents a difficult issue since it is basically NP-complete. Thus, many variants based on approximation techniques, specially those from Swarm Intelligence (SI), have been proposed. These techniques have the ability of searching for problem solutions in a very efficient way. This paper surveys SI-based job scheduling algorithms for bag-of-tasks applications (such as PSEs) on distributed computing environments, and uniformly compares them based on a derived comparison framework. We also discuss open problems and future research in the area.

An ACO-inspired algorithm for minimizing weighted flowtime in cloud-based parameter sweep experiments

Parameter Sweep Experiments (PSEs) allow scientists and engineers to conduct experiments by running the same program code against different input data. This usually results in many jobs with high computational requirements. Thus, distributed environments, particularly Clouds, can be employed to fulfill these demands. However, job scheduling is challenging as it is an NP-complete problem. Recently, Cloud schedulers based on bio-inspired techniques - which work well in approximating problems with little input information - have been proposed. Unfortunately, existing proposals ignore job priorities, which is a very important aspect in PSEs since it allows accelerating PSE results processing and visualization in scientific Clouds. We present a new Cloud scheduler based on Ant Colony Optimization, the most popular bio-inspired technique, which also exploits well-known notions from operating systems theory. Simulated experiments performed with real PSE job data and other Cloud scheduling policies indicate that our proposal allows for a more agile job handling while reducing PSE completion time.

Balancing throughput and response time in online scientific Clouds via Ant Colony Optimization (SP2013/2013/00006)

The Cloud Computing paradigm focuses on the provisioning of reliable and scalable infrastructures (Clouds) delivering execution and storage services. The paradigm, with its promise of virtually infinite resources, seems to suit well in solving resource greedy scientific computing problems. The goal of this work is to study private Clouds to execute scientific experiments coming from multiple users, i.e., our work focuses on the Infrastructure as a Service (IaaS) model where custom Virtual Machines (VM) are launched in appropriate hosts available in a Cloud. Then, correctly scheduling Cloud hosts is very important and it is necessary to develop efficient scheduling strategies to appropriately allocate VMs to physical resources. The job scheduling problem is however NP-complete, and therefore many heuristics have been developed. In this work, we describe and evaluate a Cloud scheduler based on Ant Colony Optimization (ACO). The main performance metrics to study are the number of serviced users by the Cloud and the total number of created VMs in online (non-batch) scheduling scenarios. Besides, the number of intra-Cloud network messages sent are evaluated. Simulated experiments performed using CloudSim and job data from real scientific problems show that our scheduler succeeds in balancing the studied metrics compared to schedulers based on Random assignment and Genetic Algorithms.

Simulation on cloud computing infrastructures of parametric studies of nonlinear solids problems

Scientists and engineers are more and more faced to the need of computational power to satisfy the ever-increasing resource intensive nature of their experiments. Traditionally, they have relied on conventional computing infrastructures such as clusters and Grids. A recent computing paradigm that is gaining momentum is Cloud Computing, which offers a simpler administration mechanism compared to those conventional infrastructures. However, there is a lack of studies in the literature about the viability of using Cloud Computing to execute scientific and engineering applications from a performance standpoint. We present an empirical study on the employment of Cloud infrastructures to run parameter sweep experiments (PSEs), particularly studies of viscoplastic solids together with simulations by using the CloudSim toolkit. In general, we obtained very good speedups, which suggest that disciplinary users could benefit from Cloud Computing for executing resource intensive PSEs.

Multi-objective Swarm Intelligence schedulers for online scientific Clouds

Cloud Computing is a promising paradigm for parallel computing. However, as Cloud-based services become more dynamic, resource provisioning in Clouds becomes more challenging. The paradigm, with its promise of virtually infinite resources, seems to suit well in solving resource greedy scientific computing problems. In a Cloud, an appropriate number of Virtual Machines (VM) is created and allocated in physical resources for executing jobs. This work focuses on the Infrastructure as a Service (IaaS) model where custom VMs are launched in appropriate hosts available in a Cloud to execute scientific experiments coming from multiple users. Finding optimal solutions to allocate VMs to physical resources is an NP-complete problem, and therefore many heuristics have been developed. In this work, we describe and evaluate two Cloud schedulers based on Swarm Intelligence (SI) techniques, particularly Ant Colony Optimization (ACO) and Particle Swarm Optimization (PSO). The main performance metrics to study are the number of serviced users by the Cloud and the total number of created VMs in online (non-batch) scheduling scenarios. We also perform a sensitivity analysis by varying the specific-parameter values of each algorithm to evaluate the impact on the performance of the two objective metrics. The intra-Cloud network traffic is also measured. Simulated experiments performed using CloudSim and job data from real scientific problems show that the use of SI-based techniques succeeds in balancing the studied metrics compared to Genetic Algorithms.

SI-based scheduling of scientific experiments on Clouds

Scientists and engineers usually require huge amounts of computing power for performing their experiments. Precisely, Parameter Sweep Experiments (PSE) allow these kind of users to perform simulations by running the same scientific code with different input data, which results in many CPU-intensive jobs and thus computing environments such as Clouds must be used. We describe two Cloud schedulers based on two popular swarm intelligence (SI) techniques, namely ant colony optimization (ACO) and particle swarm optimization (PSO), to allocate virtual machines (VM) to physical Cloud resources. The main performance metrics to study are the number of serviced users by the Cloud -i.e., the number of Cloud users that the scheduler is able to successfully serve- and the total number of created VMs, in dynamic (non-batch) scheduling scenarios. Simulated experiments performed by using CloudSim and real PSE job data suggest that our schedulers, through a weighted metric, perform competitively with respect to the number of serviced users and achieve an effective assignment of VMs compared to a scheduler based on Genetic Algorithms.

Meta-heuristic based autoscaling of cloud-based parameter sweep experiments with unreliable virtual machines instances

Abstract Cloud Computing is the delivery of on-demand computing resources over the Internet on a pay-per-use basis and is very useful to execute scientific experiments such as parameter sweep experiments (PSEs). When PSEs are executed it is important to reduce both the makespan and monetary cost. We propose a novel tri-objective formulation for the PSEs autoscaling problem considering unreliable virtual machines (VM) pursuing the minimization of makespan, monetary cost and probability of failures. We also propose a new autoscaler based on the Non-dominated Sorting Genetic Algorithm II able to automatically determine the right amount for each type of VM and pricing scheme, as well as the bid prices for the spot instances. Experiments show that the proposed autoscaler achieves great improvements in terms of makespan and cost when it is compared against Scaling First and Spot Instances Aware Autoscaling.

SI-based Scheduling of Parameter Sweep Experiments on Federated Clouds

Scientists and engineers often require huge amounts of computing power to execute their experiments. This work focuses on the federated Cloud model, where custom virtual machines (VM) are launched in appropriate hosts belonging to different providers to execute scientific experiments and minimize response time. Here, scheduling is performed at three levels. First, at the broker level, datacenters are selected by their network latencies via three policies –Lowest-Latency-Time-First, First-Latency-Time-First, and Latency-Time-In-Round–. Second, at the infrastructure level, two Cloud VM schedulers based on Ant Colony Optimization (ACO) and Particle Swarm Optimization (PSO) for mapping VMs to appropriate datacenter hosts are implemented. Finally, at the VM level, jobs are assigned for execution into the preallocated VMs. Simulated experiments show that the combination of policies at the broker level with ACO and PSO succeed in reducing the response time compared to using the broker level policies combined with Genetic Algorithms.

Broker Scheduler based on ACO for Federated Cloud-based scientific experiments

Federated Clouds are infrastructures arranging physical resources from different datacenters. A Cloud broker intermediates between users and datacenters to support the execution of jobs through Virtual Machines (VM). We exploit federated Clouds to run CPU-intensive jobs, in particular, Parameter Sweep Experiments (PSE). Specifically, we study a broker-level scheduler based on Ant Colony Optimization (ACO), which aims to select the datacenters taking into account both the network latencies and the availability of resources. The less the network latency, the lower the influence on makespan. Moreover, when more VMs can be allocated in datacenters with lower latency, more physical resources can be taken advantage of, and hence job execution time decreases. Then, once our broker-level scheduler has selected a datacenter to execute jobs, VMs are allocated in the physical machines of that datacenter by another intra-datacenter scheduler based on ACO. Experiments performed using CloudSim and job data from a real PSE show that our ACO-based broker-level scheduler succeeds in reducing the makespan compared to similar schedulers based on latency-aware greedy and round robin heuristics.

A Three-level Scheduler to Execute Scientific Experiments on Federated Clouds

For executing current simulated scientific experiments it is necessary to have huge amounts of computing power. A solution path to this problem is the federated Cloud model, where custom virtual machines (VM) are scheduled in appropriate hosts belonging to different providers to execute such experiments, minimizing response time. In this paper, we study schedulers for federated Clouds. Scheduling is performed at three levels. First, at the broker level, datacenters are selected by their network latencies via three policies -Lowest-Latency-Time-First, First-Latency-Time-First, and Latency-Time-In-Round-. Second, at the infrastructure level, two Cloud VM schedulers based on Ant Colony Optimization (ACO) and Particle Swarm Optimization (PSO) are implemented. At this level the scheduler is responsible for mapping VMs to datacenter hosts. Finally, at the VM level, jobs are assigned for execution into the preallocated VMs. We evaluate, through simulated experiments, how the proposed three-level scheduler performs w.r.t. the response time delivered to the user as the number of Cloud machines increases, a property known as horizontal scalability.