First Fit Decreasing Research Articles

Task Partitioning Hybrid Algorithms for Multi-Core Processors

Multi-core Processors Rapidly developing in the world we live in and Companies Manufacturers continuous efforts to manufacture faster and intelligent chips. The use of multi-core processors has become widespread in many areas; therefore, task Partitioning for multi-core processors has become a hot issue in the research community. Many types of research have been done on task Partitioning and Task Mapping from various perspectives by researchers. The existing partitioning task algorithms still have negatives, such as low usage of the processor or not making the most of the cores efficiency on the processor. However, partitioning tasks onto processors is a significant challenge among the challenges facing the multi-core processor. In this paper, to solve this challenge we proposed developing the framework that uses hybrid partitioning algorithms to mix the Blocking-Aware Algorithm with First-Fit Decreasing (FFD) algorithm, based on each of these algorithms advantages.

A Layer & Request Priority-based Framework for Dynamic Resource Allocation in Cloud- Fog - Edge Hybrid Computing Environment

One of the most promising frameworks is the fog computing paradigm for time-sensitive applications such as IoT (Internet of Things). Though it is an extended type of computing paradigm, which is mainly used to support cloud computing for executing deadline-based user requirements in IoT applications. However, there are certain challenges related to the hybrid IoT -cloud environment such as poor latency, increased execution time, computational burden and overload on the computing nodes. This paper offers A Layer & Request priority-based framework for Dynamic Resource Allocation Method (LP-DRAM), a new approach based on layer priority for ensuring effective resource allocation in a fog-cloud architecture. By performing load balancing across the computer nodes, the suggested method achieves an effective resource allocation. Unlike conventional resource allocation techniques, the proposed work assumes that the node type and the location are not fixed. The tasks are allocated based on two constrain, duration and layer priority basis i.e, the tasks are initially assigned to edge computing nodes and based on the resource availability in edge nodes, the tasks are further allocated to fog and cloud computing nodes. The proposed approach's performance was analyzed by comparing it to existing methodologies such as First Fit (FF), Best Fit (BF), First Fit Decreasing (FFD), Best Fit Decreasing (BFD), and DRAM techniques to validate the effectiveness of the proposed LP-DRAM.

Energy-efficient virtual machine placement in data centres via an accelerated Genetic Algorithm with improved fitness computation

Energy efficiency is a critical issue in data centre management, which is the foundation for cloud computing. The VM placement has a considerable impact on a data centre's energy efficiency and resource utilisation. The assignment of VMs to PMs is an NP-hard problem without an easy way to find an optimal solution, particularly in large-scale data centres. In this study, the VM placement problem is formulated as a constrained optimisation problem. The Genetic Algorithm (GA) is a suitable method for solving this problem in terms of the quality of the solution. However, GA is time-consuming to obtain an optimal solution in the large scale optimisation problem. Therefore, this paper focuses on accelerated GA for energy-efficient VM placement. As the most time-consuming element of the GA is the calculation of its fitness function, this paper simplifies this calculation through a new fitness function in GA. Simulation results of small-, medium-, and large-scale data centres demonstrate that our accelerated GA is faster than the standard GA and gives better quality of solution than the First Fit Decreasing (FFD) algorithm, respectively. The findings of our GA with the new fitness function reveal an 8% energy saving for our GA compared to FFD and a 66% reduction in our GA execution time compared to the standard GA with standard energy formula as a fitness function. The number of generations in our GA is reduced by about 50% in comparison with the standard GA. Moreover, we started with 3000 PMs in the large-scale dataset, and only 1086 PMs were actually used after running our GA. Therefore, we may switch off far more PMs for energy savings from our GA results than those from the standard GA.

Mapping Arbitrary Logic Functions onto Carry Chains in FPGAs

Current Field Programmable Gate Arrays (FPGAs) provide fast routing links and special logic to perform carry operations; however, these resources can also be used to implement non-arithmetic circuits. In this paper, a new approach for mapping logic functions onto carry chains is presented. Unlike other approaches, the proposed technique can be applied to any logic function. The presented technique includes: (1) an architecture that is composed of blocks that implement AND and OR functions (called CANDs and CORs, respectively) by means of Look-Up-Tables (LUTs) and carry-chain resources; and (2) a mapping algorithm to reduce both the delay of the critical path and the number of used FPGA resources. The algorithm uses a heuristic to interconnect CORs and CANDs in order to reduce the delay. The problem of mapping the maxterms (or minterms) of a function to LUTs has been modelled as a Set Bin Packing (SBP) problem. Since SBP is NP-Hard, a greedy algorithm has been proposed, which is based on the First Fit Decreasing (FFD) heuristic. The results obtained have been compared with the conventional technique using both speed and area optimization. For this purpose, a large synthetic set of test cases has been generated. The proposed technique improves both the speed and area results for the vast majority of functions whose conventional implementation requires more than four logic levels. It is important to highlight that the improvement of one parameter (speed or area) is not achieved at the expense of the other.

An exact comparison of global, partitioned, and semi-partitioned fixed-priority real-time multiprocessor schedulers

We evaluate the performance of global, partitioned, and semi-partitioned fixed-priority ( FP ) multiprocessor schedulers for sporadic real-time tasks. By conducting a range of exhaustive experiments, we support a prior observation of Baruah regarding the incomparability of global and partitioned fixed-priority schedulers, by demonstrating the existence of multiple task sets, which are schedulable by either one or another approach, but not both simultaneously. In addition, we demonstrate that a suboptimal semi-partitioning heuristic, named Deadline Monotonic with Priority Migration ( DM - PM ), in certain cases outperforms both global and partitioned FP schedulers. Unlike other works, our evaluation is based on an exact schedulability test for global scheduling, as well as relies on optimal task priorities for global FP , and an optimal partitioning of tasks to processors. However, due to the exponential computational complexity of the available exact schedulability test, our evaluation is constrained to a limited number of tasks in a set. On average, partitioned scheduling performs better for a lower number of heavy tasks in a task set, while global scheduling is the opposite. The schedulability gain of one approach over another might exceed 50%, for both cases. To make such an evaluation possible, we derive a method to compute optimal task priorities for global FP by using a set of pruning rules. We demonstrate that the gain of such optimal priorities over other widely used priority assignment heuristics, including Deadline Monotonic ( DM ), can be significant, in certain cases. Moreover, we evaluate the performance of an optimal task partitioning over the available suboptimal partition heuristics, namely first-fit decreasing (FFD) and worst-fit decreasing (WFD), and confirm no significant performance difference between them, for considered task settings. Software tools used in our evaluation are implemented using C++ libraries, and are publicly available.

A first Fit type algorithm for the coupled task scheduling problem with unit execution time and two exact delays

The considered coupled task problem (CTP) is to schedule n jobs, each consisting of two (sub)tasks, on a single machine. Exact delay times are between the subtasks of a job and the makespan has to be minimized. It has been proven that the problem is strongly NP-hard in general case (see Orman and Potts (1997)), even if the lengths of the subtasks are identical. This paper considers a special case of CTP where there are jobs with two different delay times only. The complexity status of this problem is unknown. We will present an algorithm – called First Fit Decreasing (FFD) – and we will prove that its approximation ratio is in the interval (1.57894,1.57916).

Network policy aware placement of tasks for elastic applications in IaaS-cloud environment

Using cloud computing as a base, new technologies like data analytics, Internet of Things, machine learning etc., have emerged. Applications that use these technologies, depend on cloud datacenters (DC) for their computing power. Performance of these applications depends on dynamic resource provisioning by DC, as there is unpredictability of rate at which data arrives for immediate processing. Cloud service providers implement this dynamism in Infrastructure-as-a-Service (IaaS) environment, using elastic virtual machines (VM). Placing these VMs onto same physical machines (PM) and/or on the network neighborhood machines is believed to increase application performance as the network latency is minimal. Deploying sub-optimal VM placement schemes creates unwanted cross network traffic resulting in poor application performance and increases the DC operating cost. This paper formulates the policy and elastic aware placement (PEAP) as an optimization problem, with additional constraints such as fixed PM, balanced PM and co-location VMs. Further, we propose PEAP algorithm which considers individual requests demanding for one or more VMs as a whole for placement along with the life-time of requests. Proposed algorithm gives optimal VM placements for increased application performance and DC efficacy. CloudSimPlus based experiments demonstrate that as compared to first fit decreasing (FFD). First fit increasing (FFI) and first come first serve (FCFS) algorithms, the proposed technique leads to reduced resource fragmentation and resource migrations. PEAP achieves placement of all the elastic VMs together with reduced network cost, thereby increasing the application performance.

Simultaneous application assignment and virtual machine placement via ant colony optimization for energy-efficient enterprise data centers

Enterprise cloud data centers consume a tremendous amount of energy due to the large number of physical machines (PMs). These PMs host a huge number of virtual machines (VMs), on which a vast number of applications are deployed. Existing research uses two separate layers to manage data center resources: application assignment to VMs, and VM placement to PMs, each of which is a bin packing problem. While this consecutive two-layer bin packing (Consec2LBP) makes the problems easier to solve, it also limits further improvement in the quality of solution. To address this issue, an integrated any colony optimization approach is proposed in this paper to deal with both layers simultaneously. It formulates the two-layer resource management into an integrated two-layer bin packing (Int2LBP) optimization problem. Then, an integrated first fit-decreasing (FFD) algorithm Int2LBP_FFD is proposed to solve this optimization problem. Using the result of Int2LBP_FFD as an initial solution, an integrated ant colony system (ACS) algorithm Int2LBP_ACS is further developed to improve the quality of solution. Simulation experiments are conducted to demonstrate the effectiveness of our integrated approach.

EMC2: Energy-efficient and multi-resource- fairness virtual machine consolidation in cloud data centres

The rapid rise in the cloud service adoption reflects the growth of Cloud Data Centers' (CDCs) number, size, energy consumption and eco-unfriendly carbon footprints. In CDCs, Virtual Machine Consolidation (VMC) plays a significant role in reducing their energy consumption and thereby reducing the carbon footprints. The state-of-the-art VMC heuristics based on First-Fit Decreasing (FFD) and Dominant Residual Resource (DRR) called DRR-FFD and DRR-BinFill are grouping the VMs based on single VM resource. We attempt to further reduce the energy consumption of CDCs through the proposed EMC2, an energy-efficient VMC framework that employs our multi-resource-fairness based VM selection heuristics, namely VMNeAR-H (Hierarchical), VM NeAR- D (Directed Hierarchical) and VM NeAR-E (Euclidean Distance). A dataset extracted from ENERGY STAR® containing the heterogeneous physical machine resource capacities and their estimated energy consumptions is utilised in the simulation experiments. The proposed EMC2-VMNeAR-D heuristic dominates the existing DRR heuristics in terms of total energy consumed by all the physical machines in the CDC (3.318 % energy savings on average of 40 schedules = 185107 kWh).

Utilization-aware energy-efficient virtual machine placement in cloud networks using NSGA-III meta-heuristic approach

One of the most important issues in the context of cloud computing concerns the placement of virtual machines (VMs). The purpose of multi-objective virtual machine placement (MO-VMP) is to find the best place of VMs on physical machines (PMs) so as to reach predetermined goals. In this regard, a fundamental goal is maximizing the utilization of available resources while minimizing energy consumption. It is clear that inefficient use of computing resources (for instance CPU, memory, storage capacity, and bandwidth) could cause increased energy wastage. On the other hand, with optimal placement of VMs on PMs, one may prevent migrating them from one PM to another in the future, itself a secondary cause of increased energy consumption. Concerning the MO-VMP, there are very serious challenges in previous studies. Some of these works have attempted to minimize the number of active PMs. Others have investigated minimizing rack link traffic and optimizing communication and VM migration costs regarding routing goals. Since the MO-VMP is an NP-hard problem and involves high spatial and temporal complexities, heuristic and meta-heuristic methods have been widely used to solve the problem in the past decade. In the present research, we use the non-dominated sorting genetic algorithm (NSGA-III) to determine the optimal MO-VMP. To this end, a multi-objective optimizing problem is designed, and after introducing a non-linear convex optimization solution, we solve it with the NSGA-III method. Our main purpose is to minimize overall resource loss while minimizing power consumption as well as decreasing the number of active PMs. The simulation results on the CloudSim simulator confirm the superiority of the proposed method over basic methods such as first-fit decreasing (FFD) and exact mathematical approaches in terms of significant criteria such as execution time, utilization, resource wastage, and energy consumption.

Hybrid Solution for Container Placement and Load Balancing based on ACO and Bin Packing

Currently, data centers energy consumption in the cloud is attracting a lot of interest. One of the most approaches to optimize energy and cost in data centers is virtualization. Recently, a new type of container-based virtualization has ap-peared, containers are considered very light and modular virtual machines, they offer great flexibility and the possibility of migra-tion from one environment to another, which allows optimizing applications for the cloud. Another approach to saving energy is to consolidate the workload, which is the amount of processing that the computer has to perform at any given time. In this article, we will study the container placement algorithm that takes into account the QoS requirements of different users in order to minimize energy consumption. Thus, we proposed a Hybrid approach for managing resources and workload based on ant colony optimization (ACO) and the first-fit decreasing (FFD) algorithm to avoid unnecessary power consumption. The results of the experiment indicate that using the first-fit decreasing algorithm (FFD) for container placement is better than ant colony optimization especially in a homogeneous systems. On the other hand the ant colony optimization shows very satisfying results in the case of workload management.

Placement Strategy for Intercommunicating Tasks of an Elastic Request in Fog-Cloud Environment

Cloud computing hosts large number of modern day applications using the virtualization concept. However, end-to-end network latencies are detrimental to the performance of IoT (Internet of Things) applications like video surveillance and health monitoring. Although edge/fog computing alleviates the latency concerns to some extent, it is still not suitable for applications having intercommunicating tasks. Further, these applications can be elastic in nature and demand more tasks during their life-time. To address this gap, in this paper a network aware co-allocation strategy for the tasks of an individual applications is proposed. After modelling the problem using bin packing approach with additional constraints, the authors propose a novel heuristic IcAPER,(Inter-communication Aware Placement for Elastic Requests) algorithm. The proposed algorithm uses the network neighborhood machine for placement, once current resource is fully utilized by the application. Using CloudsimPlus simulator the performance IcAPER algorithm is compared with First Come First Serve (FCFS), Random and First Fit Decreasing (FFD) algorithms for the parameters (a) resource utilization (b) resource fragmentation and (c) Number of requests having intercommunicating tasks placed on to same PM. Extensive simulation results shows IcAPER maps 34% more tasks on to the same PM and also increase the resource utilization by 13% while decreasing the resource fragmentation by 37.8% when compared to other algorithms in our consideration.

Special Issue on Recent Trends and Future of Fog and Edge Computing, Services and Enabling Technologies

Special Issue on Recent Trends and Future of Fog and Edge Computing, Services and Enabling Technologies

Energy-efficient Virtual Machine Allocation Technique Using Flower Pollination Algorithm in Cloud Datacenter: A Panacea to Green Computing

Cloud computing has attracted significant interest due to the increasing service demands from organizations offloading computationally intensive tasks to datacenters. Meanwhile, datacenter infrastructure comprises hardware resources that consume high amount of energy and give out carbon emissions at hazardous levels. In cloud datacenter, Virtual Machines (VMs) need to be allocated on various Physical Machines (PMs) in order to minimize resource wastage and increase energy efficiency. Resource allocation problem is NP-hard. Hence finding an exact solution is complicated especially for large-scale datacenters. In this context, this paper proposes an Energy-oriented Flower Pollination Algorithm (E-FPA) for VM allocation in cloud datacenter environments. A system framework for the scheme was developed to enable energy-oriented allocation of various VMs on a PM. The allocation uses a strategy called Dynamic Switching Probability (DSP). The framework finds a near optimal solution quickly and balances the exploration of the global search and exploitation of the local search. It considers a processor, storage, and memory constraints of a PM while prioritizing energy-oriented allocation for a set of VMs. Simulations performed on MultiRecCloudSim utilizing planet workload show that the E-FPA outperforms the Genetic Algorithm for Power-Aware (GAPA) by 21.8%, Order of Exchange Migration (OEM) ant colony system by 21.5%, and First Fit Decreasing (FFD) by 24.9%. Therefore, E-FPA significantly improves datacenter performance and thus, enhances environmental sustainability.

A Game Theoretic Approach to Estimate Fair Cost of VM Placement in Cloud Data Center

Pricing of virtual machines (VMs) with different dimensions is a challenging task. VM pricing involves both capital and operational expenditures. Capital expenditure is fixed in nature, while operational expenditure is variable. A fraction of capital expenditure is included for VM pricing. An individual pays the cloud service provider (CSP) for his requested VM. If the users cooperate among themselves they may end up paying less to CSP for their requested VMs vis-a-vis they would have paid had they requested individually. In this paper, an n -person cooperative game is adopted to determine the price that users would pay for their requested VMs under a cooperative environment. Shapley value is used to estimate the fraction of capital expenditure that would be included in the VM price. An integer linear programming is proposed for energy-efficient placement. For evaluation, VM configurations and pricing of popular CSPs—Microsoft Azure and Amazon EC2—are considered. Results show that users would pay less for their requested VMs if they cooperate. The energy consumed by the proposed VM placement technique is compared with first fit decreasing (FFD) and enhanced first fit decreasing (EFFD). It is observed that the proposed technique consumes lesser energy compared to FFD and EFFD.

Chemical reaction optimization for virtual machine placement in cloud computing

With the development of virtualization technologies, cloud data centers are faced with more and more virtual machines (VMs) requests. How to realize an efficient virtual machine placement (VMP) becomes a hot research topic. The optimal resource consumption and the utilization rate of a physical machine in the whole cloud data center can be realized by optimizing the process from the virtual machine to the physical machine. VMP is a combinatorial optimization problem which was demonstrated to be NP-hard. In this paper, a new formulation of VMP problem is presented by taking into consideration the optimization of the two following objectives: (i) minimize the energy consumption, and (ii) maximize the resource utilization. In order to achieve these targets, we propose two algorithms based on chemical reaction optimization (CRO) algorithm, namely CVP and CVV, with two types of solution representation. The proposed algorithms are compared with other optimal placement strategies, namely Cuckoo Search Optimization (CSO), Reordered Grouping Genetic Algorithm (RGGA), First Fit Decreasing (FFD) and Best Fit Decreasing (BFD). Experimental results show that the proposed CVP and CVV give better performance comparing with the other compared algorithms in terms of resource consumption and resource utilization. In term of scalability, the proposed CVV algorithm benefits from the high computational speed and performs well when there are a large number of virtual machine scheduling requests in the cloud data center.

Automotive Industry Line Board Optimization Through Operations Research Techniques

The automotive industry has suffering a lot with the current economic crisis that affects the country and, at this critical moment, more than never, they have been looking for methods to optimize their production, reducing wastes as much as possible. This paper concerns about optimization of a physical arrangement in a production line board from an automotive industry. This line board has 17 workstations (WS), each of them has been divided into 4 clusters of racks containing the parts to assembly. The goal is to get an optimized layout for these racks, which can be of different sizes and purposes, into these clusters, to maximize the area use. Those racks should accommodate assembly parts (put in boxes) in an optimized way too. This problem was treated as a Cutting and Packing Problems (CPP). In order to optimize the clusters (1st. problem), it was used a two-dimensional CPP and we proposed the use of Wang Algorithm to solve it, taking into account the demand constraints for each cluster, the racks dimensions to be assigned into them and the distances among them in order to allow the operator passage to pick up the parts to assembly. Doing these definitions, the algorithm was executed separately for each cluster and it is proposed a new layout. For the optimization of the boxes layout which are in the racks (2nd. Problem), it was a one-dimensional CPP and we proposed the algorithms First Fit (FF); First Fit Decreasing (FFD); Best Fit (BF) e Best Fit Decreasing (BFD) to solve it, in a comparatively way, in order to maximize the quantity of boxes in the racks. For that, it was considered the boxes sizes, the required quantity of each of them and the process constraints.

Package consolidation approach to the split-order fulfillment problem of online supermarkets

The split-order fulfillment problem has been a new key issue for online supermarkets. Multi-item orders to be fulfilled by multiple warehouses are split into several suborders and delivered with multiple shipments, increasing transportation cost. This paper presents a powerful order fulfillment method—package consolidation—in which two or more stock-keeping units (SKUs) in the suborders of one order will be packed into fewer packages through lateral transshipment. We propose a consolidation model and a packing model to determine which SKUs in suborders should be transshipped for package consolidation. In addition, we develop a modified first fit decreasing (FFD) algorithm to solve the packing model and a heuristic algorithm based on the breadth-first search strategy with cutting, long-sight and suborder rules to generate near optimal order fulfillment schemes in a reasonable time. The computation results verify the superiority of the package consolidation approach in solving split-order fulfillment problem. Moreover, our sensitivity analyses provide managerial insights for online supermarkets.

Simulated annealing based VM placement strategy to maximize the profit for Cloud Service Providers

Virtual machine (VM) placement strategies reported in the literature focuses mainly on minimization of power consumption and maximization of placed VMs. The revenue earned by a cloud service provider (CSP) depends on the number of VMs placed. Increasing the number of VMs placed by a CSP not only increases the power consumption but also decreases the profit margin of the CSP. In this paper, we propose a technique called maximum VM placement with minimum power consumption (MVMP) to maximize the profit earned by a CSP. The proposed technique attempts to maximize the revenue and minimize the power budget. It is formulated as a bi-objective optimization problem, and is solved using simulated annealing (SA) technique. To reach a sub-optimal solution more randomness is applied to SA. Our MVMP algorithm is compared to five state of the art algorithms in the realm of strategic VM placement, namely Marotta and Avallone (MA) approach, Hybrid genetic algorithm (HGA), Modified Best-Fit decreasing (MBFD), First-Fit decreasing (FFD) and Random deployment. We observe that MVMP performs better than Marotta and Avallone (MA) approach, HGA, MBFD, FFD and Random placement in terms of number of servers used, energy consumption, profit and execution time. Scalability of MVMP is verified using two different scenarios: (i) fixed number of VMs and, (ii) fixed number of servers. It is observed that MVMP is scalable too.

A dynamic Tenant-Defined Storage system for efficient resource management in cloud applications

Software-Defined Storage (SDS) is an evolving concept for the management of data storage from the software's perspective. Multi-tenant applications running on the cloud can benefit from the concepts introduced by SDS by managing the allocation of data storage from the tenant's perspective. A multi-tenant application should guarantee both data separation and performance isolation towards every tenant, and migration of tenant data over time should be minimized as this is both an expensive and time consuming operation. Furthermore, with cloud computing compliance with regulatory policies regarding the storage of data remains a key hurdle, as end users often have no way to specify their requirements.In this article, we present a dynamic and extensible system for the management of storage resources in multi-tenant cloud applications. In the presented approach, tenants are hierarchically clustered based on multiple scenario-specific characteristics, and allocated to storage resources using a hierarchical bin packing algorithm (static allocation). As the load changes over time, the system corresponds to these changes by reallocating storage resources when required (dynamic reallocation). We evaluate both the static and dynamic behavior of our system. Experiments confirm that the system achieves good results regarding the average bin usage, migrations over time and clustering of related tenants. On average, less than 0.01% of the total amount of data is reallocated during each migration using the dynamic Hierarchical First-Fit Decreasing (dHFFD) algorithm while achieving an average bin usage similar to First-Fit Decreasing (FFD). The dynamic Hierarchical Greedy Decreasing (dHGD) algorithm reduces the number of migrations by a factor 100 compared to dHFFD, but at the cost of provisioning additional storage instances.