A Distributed Load-balancing Scheme Based on a Complex Network Model of Cloud Servers

Kumar, Narander; Agarwal, Shalini; Zaidi, Taskeen; Saxena, Vipin

doi:10.1145/2674632.2674639

Cited by 3 publications

(5 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It then arranges these hosts ascending, according to their distance (in terms of hop count) from itself (line 10). After that, the agent assigns one to one of the sorted requests to the sorted hosts while keeping the placement constraints and the heterogeneity of the hosts (Equations ( 6), ( 12) -( 15)) into consideration (lines [13][14][15][16][17][18][19]. It is assumed that up to 95% of the capacity of each node is allocated to the requested services, and the rest is reserved for maintenance.…”

Section: Utilization Vector Cpu Utilization Memory Utilization Indexmentioning

confidence: 99%

“…In this perspective, the global objective for optimizing the placement of IoT services, i.e., MIN-VAR, aims at minimizing utilization variance among the network nodes, as a measure of load-balancing and peak-shaving. The MIN-VAR objective function, as shown in Equation (17), is a quadratic cost function [54] Sort H in terms of proximity from low to high; i, j ← 0; while (A is not empty) do Select a i from A and f j from H; if (f j satisfies the constraints based on Equations( 6), (12)(13)(14)) then 15:…”

Section: Utilization Vector Cpu Utilization Memory Utilization Indexmentioning

confidence: 99%

“…• The introduction of a model that formalizes the IoT service placement problem in a fog computing (i.e., edgeto-cloud) infrastructure, and two objectives that aim at 1 Load-balancing refers to the distribution of workload uniformly across network resources to enhance resource utilization and network efficiency [14].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Decentralized Edge-to-Cloud Load-balancing: Service Placement for the Internet of Things

Nezami,

Zamanifar,

Djemame

et al. 2020

Preprint

View full text Add to dashboard Cite

The Internet of Things (IoT) has revolutionized everyday life and expanded the scope of smart services to a broad range of domains. In ubiquitous environments, fog computing has emerged leveraging the resources in the edge-to-cloud continuum to improve the quality of service, while reducing the traffic on cloud infrastructure and networks. In such a distributed ecosystem with heterogeneous resources of various sizes and inherent dynamics such as varying service demand over time, managing resources and services is a major challenge. This paper studies two optimization objectives and formulates a decentralized load-balancing problem for IoT service placement: (global) IoT workload balance and (local) quality of service, in terms of minimizing the cost of deadline violation, service deployment, and unhosted services. The proposed solution, EPOS Fog, introduces a decentralized multi-agent system for collective learning that utilizes edge-to-cloud nodes to jointly balance the input workload across the network and minimize the costs involved in service execution. The agents locally generate possible assignments of requests to resources and then cooperatively select an assignment such that their combination maximizes edge utilization while minimizes service execution cost. Extensive experimental evaluation with realistic Google cluster workloads on various networks demonstrates the superior performance of EPOS Fog in terms of workload balance and quality of service, compared to approaches such as First Fit and exclusively Cloud-based. The findings demonstrate how distributed computational resources on the edge can be utilized more cost-effectively by harvesting collective intelligence.

show abstract

Section: Utilization Vector Cpu Utilization Memory Utilization Indexmentioning

confidence: 99%

Section: Utilization Vector Cpu Utilization Memory Utilization Indexmentioning

confidence: 99%

See 1 more Smart Citation

Decentralized Edge-to-Cloud Load-balancing: Service Placement for the Internet of Things

Nezami,

Zamanifar,

Djemame

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…If the connection time of a node exceeds the threshold, the connection will be stopped and the node will forward the tasks to another node based on regular round-robin rules. Kumar [11] proposed a static workload balance algorithm which uses ant behavior to gather information about cloud nodes so that the workload can be transmitted to the specific node. Once the request is launched, the ants and pheromones are activated and the ants start their forward paths from the ''head'' nodes.…”

Section: Related Workmentioning

confidence: 99%

A Distributed Intelligent Hungarian Algorithm for Workload Balance in Sensor-Cloud Systems Based on Urban Fog Computing

et al. 2019

View full text Add to dashboard Cite

With the help of fog computing, urban computing and intelligence novel systems can be created to improve the urban environment and the quality of human life. Sensor-cloud systems based on urban fog computing (SCS-UFC) are new intelligent network systems, which combine a cloud platform with wireless sensor networks (WSNs) as well as fog nodes to provide functions such as sensing, computation, and storage of large-scale data. Since the sensor nodes in WSNs only have limited transmission capacity, they cannot transmit their data to the cloud platform directly. Therefore, fog nodes with stronger transmission capacity are deployed to relay the data from WSNs to the cloud platform. However, different fog nodes may be burdened with different workloads (i.e., amounts of data): usually, the fog nodes with heavier workloads mean longer transmission delay and more energy consumption. If a fog node exhausts its energy, it will die and then make the network cease to work. Therefore, it is necessary to balance the workload of all fog nodes so as to reduce transmission delay and energy consumption of the sensors. However, addressing the problem is challenging because each fog node only knows local information of its neighbors, and thus it is difficult to get a global optimization result by itself. In this paper, a distributed intelligent algorithm based on the Hungarian method is proposed. First, each fog node collects the information connected with its neighboring fog nodes that are located within its transmission range. Then, a new genetic algorithm is designed to find an approximate optimization solution. Finally, each fog node decides if it should forward parts of its workload to other fog nodes so that the workloads of all fog nodes are balanced. Simulation results show that our algorithm can achieve shorter delay and less energy consumption than existing works.

show abstract

“…This paper considers QoS in terms of service execution delay and delay threshold 4. Load-balancing refers to the distribution of workload uniformly across network resources to enhance resource utilization and network efficiency[12].…”

mentioning

confidence: 99%

Decentralized Edge-to-Cloud Load Balancing: Service Placement for the Internet of Things

et al. 2021

View full text Add to dashboard Cite

The Internet of Things (IoT) requires a new processing paradigm that inherits the scalability of the cloud while minimizing network latency using resources closer to the network edge. On the one hand, building up such flexibility within the edge-to-cloud continuum consisting of a distributed networked ecosystem of heterogeneous computing resources is challenging. On the other hand, IoT traffic dynamics and the rising demand for low-latency services foster the need for minimizing the response time and a balanced service placement. Load-balancing for fog computing becomes a cornerstone for cost-effective system management and operations. This paper studies two optimization objectives and formulates a decentralized load-balancing problem for IoT service placement: (global) IoT workload balance and (local) quality of service (QoS), in terms of minimizing the cost of deadline violation, service deployment, and unhosted services. The proposed solution, EPOS Fog, introduces a decentralized multi-agent system for collective learning that utilizes edge-to-cloud nodes to jointly balance the input workload across the network and minimize the costs involved in service execution. The agents locally generate possible assignments of requests to resources and then cooperatively select an assignment such that their combination maximizes edge utilization while minimizes service execution cost. Extensive experimental evaluation with realistic Google cluster workloads on various networks demonstrates the superior performance of EPOS Fog in terms of workload balance and QoS, compared to approaches such as First Fit and exclusively Cloud-based. The results confirm that EPOS Fog reduces service execution delay up to 25% and the load-balance of network nodes up to 90%. The findings also demonstrate how distributed computational resources on the edge can be utilized more cost-effectively by harvesting collective intelligence.INDEX TERMS Agent, cloud computing, collective learning, distributed optimization, edge computing, fog computing, internet of things (IoT), load-balancing, service placement.

show abstract

A Distributed Load-balancing Scheme Based on a Complex Network Model of Cloud Servers

Cited by 3 publications

References 18 publications

Decentralized Edge-to-Cloud Load-balancing: Service Placement for the Internet of Things

Decentralized Edge-to-Cloud Load-balancing: Service Placement for the Internet of Things

A Distributed Intelligent Hungarian Algorithm for Workload Balance in Sensor-Cloud Systems Based on Urban Fog Computing

Decentralized Edge-to-Cloud Load Balancing: Service Placement for the Internet of Things

Contact Info

Product

Resources

About