Top PDF Serverless Computing and Scheduling Tasks on Cloud: A Review

Serverless Computing and Scheduling Tasks on Cloud: A Review

Serverless Computing and Scheduling Tasks on Cloud: A Review

Recently, the emergence of Function-as-a-Service (FaaS) has gained increasing attention by researchers. FaaS, also known as serverless computing, is a new concept in cloud computing that allows the services computation that triggers the code execution as a response for certain events. In this paper, we discuss various proposals related to scheduling tasks in clouds. These proposals are categorized according to their objective functions, namely minimizing execution time, minimizing execution cost, or multi objectives (time and cost). The dependency relationships between the tasks plays a vital role in determining the efficiency of the scheduling approach. This dependency may result in resources underutilization. FaaS is expected to have a significant impact on the process of scheduling tasks. This problem can be reduced by adopting a hybrid approach that combines both the benefit of FaaS and Infrastructure-as-a-Service (IaaS). Using FaaS, we can run the small tasks remotely and focus only on scheduling the large tasks. This helps in increasing the utilization of the resources because the small tasks will not be considered during the process of scheduling. An extension of the restricted time limit by cloud vendors will allow running the complete workflow using the serverless architecture, avoiding the scheduling problem.
Show more

13 Read more

Serverless computing: a multivocal literature review

Serverless computing: a multivocal literature review

Back in 2006, Amazon launched a cloud storage service: AWS S3. It provided a storage service without the need of handling the maintenance of the servers ( Adzic et al. 2017). Later on, in late 2014, a serverless platform was launched by Amazon Web Services in which computation service became serverless. And it was from this point onwards that the new paradigm of cloud revolution began (Eivy 2017). Not so long after, in 2016, other major cloud vendors released serverless platforms to compete with already existing Amazon Web Services such as Google, Microsoft and IBM (Baldini et al. 2017, C. Wolf 2017). Almost instantly, serverless computing start gaining popularity with organizations opting for the new technology to transform their businesses (Kiriaty 2016). Currently, various commercial and open source serverless services are made available such as AWS Lambda, Google cloud functions, Microsoft Azure functions, IBM cloud functions (Baldini et al. 2017 ,Persson et al. 2017) and OpenLambda is an open source initiative to provide serverless (Varghese & Buyya, 2018). Subsequently, providers are working progressively to provide the best functionality and services to differentiate themselves from their competitors (Eivy 2017 ). However, the market is mainly captured by the Amazon Web Services as they already had a strong foothold before the competitors entered the industry ( Crane et al. 2017). According to Markets and Markets report, the serverless market will grow from a $1.88 billion market in 2016 to $7.72 billion by 2021 and 80 percent by 2020 ( Leitersdorf et al. 2017, Contino 2017).
Show more

13 Read more

A Review on Optimization Techniques Used in Cloud Scheduling

A Review on Optimization Techniques Used in Cloud Scheduling

The author takes into account the use of genetic algorithms in this section. Scheduling is a vital issue in cloud computing in order to facilitate optimized usage of resources. Genetic algorithms are used in optimization techniques as they are inspired from evolutionary ideas of natural evolution. The cloud scheduling optimization problem is modeled as a population of candidate solution. Genetic algorithms can be applied for benefitting the fittest candidates. The study refers to variants of genetic algorithm for efficiently allocating tasks to resources. It is concluded that Multiple Priority Queues Genetic algorithm is an efficient method for scheduling tasks. It offers minimum make span and minimum overall costs.
Show more

5 Read more

A Brief Review of Scheduling Algorithms in Cloud Computing

A Brief Review of Scheduling Algorithms in Cloud Computing

Dr. Amit Agarwal et al. (2014) Cloud computing is an emerging technology in distributed computing which facilitates pay per model as per user demand and requirement. Cloud consists of a collection of virtual machine which includes both computational and storage facility. The primary plan of cloud computing is to provide efficient access to remote and geographically distributed resources. Cloud is mounting day by day and faces many challenges, one of them is scheduling. Scheduling refers to a set of policies to direct the order of work to be performed by a computer system. A good scheduler adapts its scheduling approach according to the changing environment and the type of task. In this research paper we presented a Generalized Priority algorithm for competent execution of task and comparison with FCFS and Round Robin Scheduling. Algorithm should be veteran in cloud Sim toolkit and result shows that it gives better performance compared to other traditional scheduling algorithm [8].
Show more

5 Read more

Scheduling of Heterogeneous Tasks in Cloud Computing using Multi Queue (MQ) Approach

Scheduling of Heterogeneous Tasks in Cloud Computing using Multi Queue (MQ) Approach

rapid increment, the requirement of computations is also increasing in cloud environment. There are multiple issues that exist in cloud environment like quality of services (QoS) requirement, minimum energy consumption and scheduling of tasks. Number of task scheduling algorithms exist in cloud computing, which schedule the tasks to available resources in easy way. In this paper, Multi Queue (MQ) task scheduling algorithm has been purposed to improve the performance of system. Multi Queue (MQ) scheduling algorithm overcomes the drawbacks of existing Round Robin and Weighted Round Robin algorithms. CloudSim toolkit has been used to simulate the proposed work. Experiment results show that the proposed Multi Queue (MQ) scheduling algorithm performs better as compared to exiting Round Robin (RR) and Weighted Round Robin (WRR) algorithms.
Show more

7 Read more

Tasks Distribution Strategy based on Cluster in MWfSCC

Tasks Distribution Strategy based on Cluster in MWfSCC

Workflow system built on high performance computing infrastructures such as p2p and grid computing are often applied to support the process automation of large scale workflow applications. MWfSCC is not built from scratch but from its predecessor GoMWfS (goal oriented migrating workflow system). The practical applications show the simplicity and validity of our prototype system in modeling and executing cloud computing transactions. Traditional scheduling strategy usually emphasize on the data place, not involving the tasks distribution.

8 Read more

BAT ALGORITHM FOR ROUGH SET ATTRIBUTE REDUCTION

BAT ALGORITHM FOR ROUGH SET ATTRIBUTE REDUCTION

Resources in cloud computing environment are heterogeneous. To integrate the resources effectively and schedule users’ tasks is the key problem. The goal of resource scheduling is to coordinate various resources and manage them reasonably, optimizing resource scheduling according to task demands submitted by users. Resource scheduling consider- ing QoS constraints from the aspect of user and system load balance can make both of them be satisfied. Quality service is a measure of cloud users’ needs [8], QoS model is a extending vector, it can be described from many aspects such as complete time, cost, extension, throughput etc. They estimate QoS from different aspects. In this paper, multiple QoS constrains can be divided into three parts, they are complete time, cost and load balance. The front two parts are constraint indexes of user, the third one is system constraint index.
Show more

6 Read more

Evolutionary Algorithm Based Multi-Objective Tasks Scheduling Algorithm in Cloud Computing

Evolutionary Algorithm Based Multi-Objective Tasks Scheduling Algorithm in Cloud Computing

In today’s world, cloud computing is the hottest emerging area in field of Information technology [1]. Cloud computing is a service based, on-demand, pay per use model consisting of an inter-connected and virtualizes resources delivered over internet. In the cloud platform, task scheduling is the most important concern that aims to ensure that user’s requirement are properly and correctly satisfied by cloud infrastructure. Basically, scheduling is the process of mapping or assigning task to the available resources after looking the characteristics of task. An efficient scheduling mechanism should meet user’s requirement and helps service provider to provide good quality of service, so as to enhance the overall system performance.
Show more

6 Read more

Abstract: Task scheduling plays the key role systems in cloud computing. Scheduling of tasks cannot be

Abstract: Task scheduling plays the key role systems in cloud computing. Scheduling of tasks cannot be

Task scheduling is one of the most well-known problems in cloud computing so; there is always a chance of amendment of previously completed work in this specific field. The researchers at their own time performed their work according to their knowledge space and after some interval their work had been carried out some other people. During scheduling they had well thought-out various techniques and applied constraints but as the cloud computing is too vast that they had not been able to capture the all aspects at same time but they mentioned these facts that there is a chance of amendment of algorithms and which part has to be modified.
Show more

7 Read more

A Review on Task Model and Task Scheduling in Cloud Computing

A Review on Task Model and Task Scheduling in Cloud Computing

The HEFT Algorithm is one of the best and well accepted list deployed heuristics. The HEFT calculation is a compelling answer for the DAG Scheduling issue on heterogeneous framework. The confinement of HEFT calculation is that it utilizes procedures that are all static methodologies of the mapping issue that accept static conditions for a given time. HEFT is a two- phase scheduling algorithm with the heterogeneous processors. The phase which is used to assign the priority to tasks is the first phase, known as the task prioritization phase and for assigning the priority, the upward rank of every task is calculated. The upward rank is the critical path of the task that is the maximum amount of communication time and the average implementation time right from the start of any task to the end of any task. The phase which is used to schedule the tasks into the process which provides the EFT of the task is the Processor Selection phase (second phase). This phase has the insertion based
Show more

6 Read more

MULTIPLE VIRTUAL MACHINE IN A CUMULATIVE SERVER USING XEN BALLOON DRIVER BY IMPLEMENTING REAL TIME TASK ORIENTED TASK SCHEDULING ALGORITHM

MULTIPLE VIRTUAL MACHINE IN A CUMULATIVE SERVER USING XEN BALLOON DRIVER BY IMPLEMENTING REAL TIME TASK ORIENTED TASK SCHEDULING ALGORITHM

A system for automatic memory control based on the balloon driver in VMs. Researchers can download our toolkit. The project aims to optimize the running times of applications in consolidated environments by overbooking and/or balancing the memory pages of VMs. The system is lightweight and can be completely integrated into user space without interfering with VMM operation. Design a global-scheduling algorithm based on the dynamic baseline to determine the optimal allocation of memory globally. In this existing system evaluate optimized solution to memory allocation using real workloads that run across VMs. The virtualization technique enables multiple virtual machines (VMs) to be placed on the same physical hosts and supports the live migration of VMs between physical hosts based on the performance requirements. When VMs do not use all the provided resources, they can be logically resized and consolidated to the minimum number of physical hosts, while idle nodes can be switched to sleep or hibernate mode to eliminate the idle energy consumption and thus reduce the total energy consumption in cloud data centers. Cloud can achieve the same level of computing power as a supercomputer does, but at a much reduced cost. Cloud is like a virtual supercomputer. However, need to consider about many conditions such as network status and resource status because the members of Cloud are connected by networks. Cloud is also a heterogeneous system. Scheduling independent tasks on it is more complicated. In order to utilize the power of Cloud computing completely, need an efficient task scheduling algorithm to assign tasks to resources. This paper focuses on the efficient task scheduling EASJSA considering the completion time of tasks in a cloud environment.
Show more

6 Read more

A Review on Virtual Machine Scheduling in Cloud Computing

A Review on Virtual Machine Scheduling in Cloud Computing

C. Reddy [7] explain use of gang scheduling algorithm in cloud computing responsible for selection of best suitable resources for task execution, by taking some static and dynamic parameters and restrictions of VM into the considerations. Gang scheduling is a scheduling algorithm for parallel system that scheduled related VM to run simultaneously on different machines. Gang Scheduling is an efficient job scheduling algorithm for time sharing, already applied in parallel and distributed systems. Gang scheduling can be effectively applied in a Cloud Computing environment both performance-wise and cost-wise. Gang scheduling is a special case of job scheduling that allows the scheduling of such virtual Machines. Gang scheduling is a special case of scheduling parallel jobs in which tasks of jobs need to communicate very frequently. Gang scheduling involves high overhead since network status must be saved and then be restored when switching between tasks.
Show more

6 Read more

Cloud Computing Online Scheduling

Cloud Computing Online Scheduling

Abstract: - Cloud computing has gained a lot of attention to be used as a computing model for a variety of application domains. Task scheduling is the fundamental issue in this environment. To utilize cloud efficiently, a good task scheduling algorithm is needed to assign tasks to resources in cloud. Cloud task can be divided into two categories such as on-line mode service and the batch mode service. In this paper, online cloud task scheduling based on virtual machine adaptive fault tolerance and load balancing using ant colony algorithm is proposed. The main contribution of this work is that load balancing factor is added and the system tolerates the faults by tacking the decision on the basis of reliability of the virtual machines in scheduling process. The proposed scheduling strategy was simulated using the Cloudsim toolkit package. Experimental results show that the proposed algorithm achieved the better load balance than Join-shortest-queue (JSQ) and Modified Ant Colony Optimization (MACO) algorithms.
Show more

11 Read more

Hybrid Starling Social Spider Algorithm for Energy and Load Aware Task Scheduling in Cloud Computing

Hybrid Starling Social Spider Algorithm for Energy and Load Aware Task Scheduling in Cloud Computing

while scheduling the tasks to cloud resources[11]. During the schedule of user tasks, load and energy of each resource needs to be realized and decisions can be taken to assign a task to the best VM so that the overall makespan, degree of imbalance and energy consumption is minimized[12]. Thus, makespan optimization with load balancing is a critical issue in achieving high performance in cloud computing. Additionally, reducing the energy consumption is also an important parameter that needs to be considered while assigning the task to the resources since it sustains the whole cloud computing system with green ambience[13]. The task scheduling policies plays an important role on reducing energy consumption in order to allocate tasks on distributed VMs. In recent studies carried out by [14] have estimated in average of 55% of energy utilized by the computing system in cloud computing. Thus, green computing is also an important issue for ensuring the sustainability of cloud computing system in terms of cost and environment safety. This paper proposes a hybrid algorithm for energy efficient load balance aware task scheduling in cloud computing inspired by social spider’s foraging behavior and starling birds flock behavior. The social spider algorithm is an algorithm inspired by foraging behavior of social spiders has been adopted by [15] for global optimization. It is an intelligent exploitation of foraging strategies present in the spider’s web. This swarm intelligence helps the spiders in the population to find the food sources which is hybridized with the intelligent flocking behavior[16] of starling birds as major component for scheduling the cloud user tasks in energy efficient way. The flocks of starling birds are an interesting group behavior where simple rules of interaction among the birds sufficient to produce the collective behavior. The starling flock intelligence is incorporated to the social spider foraging in order to achieve energy efficient task scheduling in cloud computing with balanced load on VMs[17]. The remainder of this paper is organized as follows. Section II describes the various research study related to improve the energy efficiency of task scheduling cloud computing and Section III presents the background concepts of proposed system. Section IV describes the task scheduling model and detailed procedure of proposed system. Section V provides the simulation results and performance evaluation with existing techniques.
Show more

8 Read more

TASK SCHEDULING IN CLOUD COMPUTING

TASK SCHEDULING IN CLOUD COMPUTING

synchronization is, the weaker the impact it brings Task assignment involves in setting task classification according to PRI. A task that has a higher execution priority has higher PRI. Reaching the uptime or task threshold means that the time threshold of task running or the number of tasks that are waiting in the line is reached. It includes two conditions. There are two types of task threshold. One is the number of tasks waiting to be done in the queue on one virtual machine. The other one is the number of tasks that have been finished on another virtual machine. If both numbers were larger than the threshold value at the same time, these two virtual machines would be synchronized. And their tasks will be balanced and will continue working.Task equilibrium means that if there is at least one idle virtual machine and at least one overload virtual machine, other virtual machines will execute tasks independently. Xian’s paper [23] compares between dynamic scheduling algorithm based on threshold and virtual machines with the static independent job scheduling algorithm on CloudSim platform. The result suggests that when there are a fairly large number of tasks, the former can complete task allocation efficiently and reduce the running time greatly. It shows an obvious advantage over the latter.[8]
Show more

5 Read more

Scheduling Algorithms in Cloud Computing

Scheduling Algorithms in Cloud Computing

Cloud Computing has come to be perception for large scale of distributed computing and parallel processing. Cloud computing is a form of internet based computing that provides shared computer processing resources and data to computers and other devices on demand. The execution and suitability of cloud computing services always depends upon the completion of the user tasks affirmed to the cloud system. Task scheduling is one of the main types of scheduling performed. Scheduling is the major issue in establishing cloud computing system. The scheduling algorithms should order the jobs in a way where balance between improving the performance and quality of service and at the same time maintaining the efficiency and fairness among the jobs. This paper aims at studying various scheduling methods. A good scheduling technique also helps in proper and efficient utilization of the resources. Many scheduling techniques have been developed by the researchers like GA (Genetic Algorithm), PSO (Particle Swarm Optimization), Min-Min, Max-Min, Priority based Job Scheduling Algorithm
Show more

6 Read more

An Analysis of Priority, Length, and Deadline Based Task Scheduling Algorithms in Cloud Computing

An Analysis of Priority, Length, and Deadline Based Task Scheduling Algorithms in Cloud Computing

Generalized Priority Based Algorithm (GPA) [9]: The GPA mainly focuses on minimizing the completion time of tasks. In this algorithm tasks and Virtual Machines (VM). The VMs are prioritized according to million instructions per second (MIPS) VM can execute and tasks are prioritized according to length or size of tasks. The VM having highest MIPS value and the task having largest size has the highest priority. Tasks having highest priority is scheduled on the VM having the highest priority. The results of the algorithm are analyzed with basic FCFS and RR algorithms, where the outcome of GPA is better than FCFS and RR.
Show more

5 Read more

PGA Scheduling with Cloud Computing

PGA Scheduling with Cloud Computing

Review of parallel algorithm classification is the basic requirement for the computation of the tasks scheduling. Different techniques based on their characteristics anlyize the scheduling critera. It depends upon the setting of the parallel environment and the availability of the information[6][7][9][11]. A huge task efficiently sub divided into a set of sub tasks having a particular format of appropriate grain size. This strategy is based on the idea of execution of these sub tasks on parallel computing system. Such type of an abstract model having the divided parts of sub tasks that can be represented by a natural and fast showing system called DAG. In other words it is called as Directed Acyclic Graph [12][13]. A precedence relationsip created among these sub tasks and can be scheduled in a particular format is the major achivement in this deterministic scheduling problem. A deterministic scheduling problem [14] is one in which all information about the tasks and the relation to each other such as execution time and precedence relation are known to the scheduling algorithm in advance and the processor environment is hetrogeneous[15][16]. The processors having different speed or and processing capabilities is the major achivemnt of heterogenity of processors. To compute and evaluate such type of system there be need of anlysis and to discuss the task scheduling problem in such type of environemnt. To determine the probabaility of occurance of a particular job in the heterogeneous parallel computing environemnet with the help of cloud computing is the basic need. To minimize the total task finish time (execution time + waiting time or idle time) i.e. minimize the makespan is the major goal in such type of system. Here a computing environment can takes place having the parallel multiprocessor with a set of x hetrogeneous machines and these are set under a cloud:
Show more

6 Read more

Greedy scheduling of tasks with time constraints for energy-efficient cloud-computing data centers

Greedy scheduling of tasks with time constraints for energy-efficient cloud-computing data centers

Figure 9 shows the histogram of the total (cumula- tive) task response time for the MESF and random task- scheduling schemes. We use a generic time unit, which can be converted to μs, or any other specific unit. The figure shows that the random-based scheme has a smaller total task response time than the proposed scheme. This is expected as the random-based scheme selects a larger number of servers to process the tasks. The larger devi- ation in the distribution of the response time for the proposed scheme may be caused by the queueing dynam- ics because more tasks are queued for process. The small deviation for the random-based scheme indicates a small fluctuation in queue length where the queue occupancy of a server is small (or zero). The random-based scheme keeps the response time small but at the expense of using a large number of servers on active state and there- fore, a large energy consumption. On the other hand, the MESF scheduling algorithm uses longer processing time as the algorithm attempts to use the smallest number of servers, and in turn, minimizes the amount of energy con- sumed. More importantly, although not obviously shown in this figure, the task response time of the proposed task-scheduling scheme is bounded.
Show more

14 Read more

REVIEW OF TASK SCHEDULING METHODS FOR REAL TIME TASKS IN CLOUD ENVIRONMENT

REVIEW OF TASK SCHEDULING METHODS FOR REAL TIME TASKS IN CLOUD ENVIRONMENT

Cloud Computing is a type of Internet model that enables convenient, on-demand resources that can be used rapidly and with minimum effort. Cloud Computing can be IaaS, PaaS or SaaS. Scheduling of these tasks is important so that resources can be utilized efficiently with minimum time which in turn gives better performance. Real time tasks require dynamic scheduling as tasks cannot be known in advance as in static scheduling approach. There are different task scheduling algorithms that can be utilized to increase the performance in real time and performing these on virtual machines can prove to be useful. Here a review of various task scheduling algorithms is done which can be used to perform the task and allocate resources so that performance can be increased.
Show more

5 Read more

Show all 10000 documents...