Online Parallelized Service Function Chain Orchestration in Data Center Networks

Abstract

In recent years, much attention has been focused on deploying service function chains (SFCs), each of which is composed of a set of virtual network functions (VNFs) in a specified order. This is a promising approach for enabling cloud service providers to deploy user service requests more flexibly while saving costs. However, less effort has been directed toward meeting heterogeneous needs, such as high throughput or low latency of user service requests with heterogeneous bandwidth demands, especially in data center networks (DCNs). In this paper, we propose an efficient orchestration algorithm for online SFC requests. It first splits a large flow into a number of subflows and replicates the same number of sub-SFCs. Each subflow is redirected to one of these ''parallelized'' sub-SFCs, which is termed a sub-user request. Then, each sub-user request is deployed based on a worst-fit strategy, and VNFs in the same SFC are instantiated on the same server to the greatest possible extent. Our algorithm is expected to enable network load balancing, reducing the delay experienced by small flows while improving the acceptance ratio for user requests. Finally, the simulation results show that the proposed algorithm outperforms other comparable algorithms.INDEX TERMS Service function chain, splitting, data center network, orchestration, load balancing.

Low-Latency and Resource-Efficient Service Function Chaining Orchestration in Network Function Virtualization

“…If 4 , B = 1, the -th VNF 4 is deployed on the physical node B , otherwise 4 , B = 0. Formulas (10) and (11) guarantee that a VNF can be deployed on only one physical network node and that a physical network node only hosts one VNF for an SFC. Formula (12) ensures that the request resources of all VNFs deployed on the node B can not exceed all resources of this node.…”

“…The computing resources requested by each VNF follow a uniform distribution, U (10,20). The bandwidth resources requested by each virtual network link are subject to a uniform distribution, U (10,20). For fairness, we use the same network topology and parameters when comparing the two algorithms.…”

“…With the expansion of the network scale and the increase in SFC requests, ensuring successful SFC deployment is a considerable challenge. Many studies have shown that the SFC deployment problem is an NP-hard problem [9][10][11]. There is no polynomial time algorithm to solve the problem.…”

A Survey of Multiobjective Evolutionary Algorithms Based on Decomposition: Variants, Challenges and Future Directions

“…Sun et al [108] designed a heuristic algorithm to realize low-latency and resourceefficient SFC orchestration. Similar to the idea of decomposition in MOEA/D, Sun et al [109] split a large flow into a number of subflows and each subflow was redirected to one of these ''parallelized'' sub-SFCs. Xu et al [110] considered the requested resource quantity of each virtual network function (VNF) and VNFs precedence, which means that, for an SFC, its one VNF must be executed before the special VNF.…”

Deadline-Aware Fast One-to-Many Bulk Transfers over Inter-Datacenter Networks

“…In order to guarantee quality and highly available services, these services are generally operated on top of multiple datacenters dispersed in different cities or countries close to regional users [1], [2], [3]. Given such architecture, it is necessary to replicate data, such as updated machine learning models and multimedia, across the datacenters that offer the same services to reduce the response latency to access services, enhance failure recovery, and expedite geo-distributed data analytics [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16]. A representative example is the search service, where the search engines periodically synchronize their search index databases across multiple locations to improve overall search quality (e.g., relevance and precision) and user satisfaction [9], [10].…”