Towards self-adaptive bandwidth allocation for low-latency communications with reinforcement learning

Ruan, Ling; Dias, Maluge Pubuduni Imali; Wong, Elaine

doi:10.1016/j.osn.2020.100567

Cited by 10 publications

(6 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Each physical node's amount of computing resources are randomly set in a discrete-uniform range [20,100] gigabit (Gb). Each physical link has a bandwidth in a discrete-uniform range [5,10] gigabit per second (Gbps). The propagation delay of a physical link is randomly set in a uniform range [10,50] microseconds (µs).…”

Section: A Simulation Settingsmentioning

confidence: 99%

“…Each physical link has a bandwidth in a discrete-uniform range [5,10] gigabit per second (Gbps). The propagation delay of a physical link is randomly set in a uniform range [10,50] microseconds (µs). In P-NSR, the number of SF nodes is randomly set in a discrete-uniform range [5,20], edge density of PG is randomly set in a uniform range [0, 1], and bandwidth demand is random in a discrete-uniform range [1,5] Gbps.…”

Section: A Simulation Settingsmentioning

confidence: 99%

“…In P-NSR, the number of SF nodes is randomly set in a discrete-uniform range [5,20], edge density of PG is randomly set in a uniform range [0, 1], and bandwidth demand is random in a discrete-uniform range [1,5] Gbps. Each SF node requires computing resources in a discreteuniform range [10,25] Gb and needs the maximum processing delay in a uniform range [10,30] µs. The service source and destination of the P-NSR are randomly selected from the PN.…”

Section: A Simulation Settingsmentioning

confidence: 99%

“…Recently, NFV techniques are applied in 5G networks to facilitate the low-lantecy service delivery [6]- [9], where 5G technologies are designed to significantly reduce latency (as much as 10x) [10]- [12]. Under such scenarios, the total processing delay from the serially-running SFs in an SFC may be comparable to the propagation delay and could be the bottleneck of optimizing the overall end-to-end latency for SFC delivery [13], [14].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Towards Optimal Parallelism-Aware Service Chaining and Embedding

Zheng

Shen

Cao

et al. 2022

IEEE Trans. Netw. Serv. Manage.

View full text Add to dashboard Cite

Emerging 5G technologies can significantly reduce end-to-end service latency for applications requiring strict quality of service (QoS). With network function virtualization (NFV), to complete a client's request from those applications, the client's data can sequentially go through multiple service functions (SFs) for processing/analysis but introduce additional processing delay. To reduce the processing delay from the serially-running SFs, network function parallelism (NFP) that allows multiple SFs to run in parallel is introduced. In this work, we study how to apply NFP into the SF chaining and embedding process such that the latency, including processing and propagation delays, can be jointly minimized. We introduce a novel augmented graph to address the parallel relationship constraint among the required SFs. Considering parallel relationship constraints, we propose a novel problem called parallelism-aware service function chaining and embedding (PSFCE). For this problem, we propose a near-optimal maximum parallel block gain (MPBG) first optimization algorithm when computing resources at each physical node are enough to host the required SFs. When computing resources are limited, we propose a logarithm-approximate algorithm, called parallelism-aware SFs deployment (PSFD), to jointly optimize processing and propagation delays. We conduct extensive simulations on multiple network scenarios to evaluate the performances of our schemes. Accordingly, we find that (i) MPBG is near-optimal, (ii) the optimization of end-to-end service latency largely depends on the processing delay in small networks and is impacted more by the propagation delay in large networks, and (iii) PSFD outperforms the schemes directly extended from existing works regarding end-to-end latency.

show abstract

Section: A Simulation Settingsmentioning

confidence: 99%

Section: A Simulation Settingsmentioning

confidence: 99%

Section: A Simulation Settingsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Towards Optimal Parallelism-Aware Service Chaining and Embedding

Zheng

Shen

Cao

et al. 2022

IEEE Trans. Netw. Serv. Manage.

View full text Add to dashboard Cite

show abstract

“…Thus, in this study, the SI is improved to cover longer distance of an LR-PON, concerning the least mean delay during bandwidth allocation process among traffic transmission containers (T-CONTs). There are also other works on LR-PON which focuses on polling and scheduling algorithm such as in [14,21,27,[30][31][32][33][34] but most of them are on E-PON and some are not GPON-FSAN compliant.…”

Section: Related Work On Dba Algorithmsmentioning

confidence: 99%

Dynamic bandwidth allocation algorithm for long reach passive optical network

2021

View full text Add to dashboard Cite

Next generation broadband access networks are gaining more interests from many key players in this field. The demands for longer reach and higher bandwidth are among the driving factors for such network as it can reach wider area up to 100 km, even beyond; has enhanced bandwidth capacity and transmission speed, but with low cost and energy consumption. One promising candidate is long reach passive optical network, a simplified network with reduced number of network elements, equipment interfaces, and even nodes; which leads to a significant reduction in the network's capital expenditure and operational expenditure. Outcome of an extended reach often results in increased propagation delay of dynamic bandwidth allocation messages exchange between the optical line terminals and optical network units, leading to the degradations of bandwidth allocation and quality of service support. Therefore, an effective bandwidth allocation algorithm with appropriate service interval setup for a long reach network is proposed to ensure the delay is maintained under ITU-T G.987.1 standard requirement. An existing algorithm is improved in terms of service interval so that it can perform well beyond 100 km. Findings show that the improved algorithm can reduce the mean delay of high priority traffic classes for distance up to 140 km.

show abstract

Low-latency resource elements scheduling based on deep reinforcement learning model for UAV video in 5G network

Jiang

Qiu

et al. 2021

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

We consider the problem of resource elements allocation in a network environment with multiple users. Previous studies have done a lot of works using traditional methods in terms of bandwidth allocation, which is sufficient to serve for 4G network. However, it cannot be neglected to provide more efficient and intelligent scheduling policies in haste, due to growing demands on high resolution video and image transmission in 5G network. To fit the condition taking resource elements as scheduling unit in 5G network, we proposed deep Q network (DQN) algorithm based on the requirement of low time latency and high resource utilization rate to solve resource elements (RE) scheduling problem. Ultimately, we give out the optimal allocation scheme of resource elements (RE) for four users in fixed condition of time latency and resource utilization rate.

show abstract

Towards self-adaptive bandwidth allocation for low-latency communications with reinforcement learning

Cited by 10 publications

References 15 publications

Towards Optimal Parallelism-Aware Service Chaining and Embedding

Towards Optimal Parallelism-Aware Service Chaining and Embedding

Dynamic bandwidth allocation algorithm for long reach passive optical network

Low-latency resource elements scheduling based on deep reinforcement learning model for UAV video in 5G network

Contact Info

Product

Resources

About