Resource Reservation within Sliced 5G Networks: A Cost-Reduction Strategy for Service Providers

Monteil, Jean-Baptiste; Hribar, Jernej; Barnard, Pieter; Li, Yong; DaSilva, Luiz A.

doi:10.1109/iccworkshops49005.2020.9145374

Cited by 15 publications

(20 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When the availability of best-effort resources is limited, due to heavy competition in the network, the model will at most reserve the tenant's own future demand, or else the minimum portion of it that cannot be covered by the resources available on a best-effort basis. 3 The outcome of this simulation produces the ground-truth labels and corresponding signals s t,k , u t,k , b t,k , d t,k , and z t,k , which our tenant receives when making reservations under this policy. We note that for the purpose of this simulation, we assume that the other remaining tenants are perfect predictors of their own traffic, and thus make reservations according to r t,i = s t,i , i ∈ K \ {k}, and also set the maximum capacity of the network to be equal to the 80th percentile of the total traffic occurring during the training data, i.e., c = P 80 ( i∈K s t,i ).…”

Section: A Deep Learning Solutionmentioning

confidence: 99%

“…In this paper, we examine the task of optimising the efficiency of a slice by means of performing short-term resource reservation (STRR) from the slice tenant's perspective. In contrast to our previous work [3], which looks at the accuracy trade-offs of various artificial intelligence (AI) models for STRR, this paper explores the explainability of AI models for STRR. In particular, our work is motivated by the fact that many AI models, such as deep neural networks (DNNs), are commonly perceived as "black-boxes" due to a lack of transparency in their behaviour [4].…”

Section: Introductionmentioning

confidence: 99%

“…Although other reservation policies are also feasible in this context, such as[3], the choice of policy is arbitrary and does not affect the primary concern of our work, which is to provide explainability to the underlying model 4. We do not include the tenant's delivered traffic as a feature, as this becomes redundant to the model once the total and undelivered amounts are given.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Resource Reservation in Sliced Networks: An Explainable Artificial Intelligence (XAI) Approach

Barnard¹,

Macaluso²,

Marchetti³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

The growing complexity of wireless networks has sparked an upsurge in the use of artificial intelligence (AI) within the telecommunication industry in recent years. In network slicing, a key component of 5G that enables network operators to lease their resources to third-party tenants, AI models may be employed in complex tasks, such as short-term resource reservation (STRR). When AI is used to make complex resource management decisions with financial and service quality implications, it is important that these decisions be understood by a human-in-the-loop. In this paper, we apply state-of-the art techniques from the field of Explainable AI (XAI) to the problem of STRR. Using real-world data to develop an AI model for STRR, we demonstrate how our XAI methodology can be used to explain the real-time decisions of the model, to reveal trends about the model’s general behaviour, as well as aid in the diagnosis of potential faults during the model’s development. In addition, we quantitatively validate the faithfulness of the explanations across an extensive range of XAI metrics to ensure they remain trustworthy and actionable.

show abstract

Section: A Deep Learning Solutionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Resource Reservation in Sliced Networks: An Explainable Artificial Intelligence (XAI) Approach

Barnard¹,

Macaluso²,

Marchetti³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Introductionmentioning

confidence: 99%

Resource Reservation in Sliced Networks: An Explainable Artificial Intelligence (XAI) Approach

Barnard

Macaluso

Marchetti

et al. 2022

ICC 2022 - IEEE International Conference on Communications

Self Cite

View full text Add to dashboard Cite

The growing complexity of wireless networks has sparked an upsurge in the use of artificial intelligence (AI) within the telecommunication industry in recent years. In network slicing, a key component of 5G that enables network operators to lease their resources to third-party tenants, AI models may be employed in complex tasks, such as short-term resource reservation (STRR). When AI is used to make complex resource management decisions with financial and service quality implications, it is important that these decisions be understood by a human-in-the-loop. In this paper, we apply state-of-theart techniques from the field of Explainable AI (XAI) to the problem of STRR. Using real-world data to develop an AI model for STRR, we demonstrate how our XAI methodology can be used to explain the real-time decisions of the model, to reveal trends about the model's general behaviour, as well as aid in the diagnosis of potential faults during the model's development.In addition, we quantitatively validate the faithfulness of the explanations across an extensive range of XAI metrics to ensure they remain trustworthy and actionable.

show abstract

“…In [15], the authors consider a hybrid reservation/spot market where the SP reservations are decided by solving a stochastic problem, and a similar mixed time scale reservation model was studied in [16]. Our previous work [17] employed demand and price predictions (via neural networks) to assist the SP's reservations, while [18] focused on slice reconfiguration costs.…”

Section: Introductionmentioning

confidence: 99%

No-Regret Slice Reservation Algorithms

Monteil

Iosifidis

DaSilva

2021

ICC 2021 - IEEE International Conference on Communications

Self Cite

View full text Add to dashboard Cite

Emerging network slicing markets promise to boost the utilization of expensive network resources and to unleash the potential of over-the-top services. Their success, however, is conditioned on the service providers (SPs) being able to bid effectively for the virtualized resources. In this paper we consider a hybrid advance-reservation and spot slice market and study how the SPs should reserve slices in order to maximize their performance while not exceeding their budget. We consider this problem in its general form, where the SP demand and slice prices are time-varying and revealed only after the reservations are decided. We develop a learning-based framework, using the theory of online convex optimization, that allows the SP to employ a no-regret reservation policy, i.e., achieve the same performance with a hypothetical policy that has knowledge of future demand and prices. We extend our framework for the scenario the SP decides dynamically its slice orchestration, where it additionally needs to learn which resource composition is performancemaximizing; and we propose a mixed-time scale scheme that allows the SP to leverage any spot-market information revealed between its reservations. We evaluate our learning framework and its extensions using a variety of simulation scenarios and following a detailed parameter sensitivity analysis.

show abstract

Resource Reservation within Sliced 5G Networks: A Cost-Reduction Strategy for Service Providers

Cited by 15 publications

References 10 publications

Resource Reservation in Sliced Networks: An Explainable Artificial Intelligence (XAI) Approach

Resource Reservation in Sliced Networks: An Explainable Artificial Intelligence (XAI) Approach

Resource Reservation in Sliced Networks: An Explainable Artificial Intelligence (XAI) Approach

No-Regret Slice Reservation Algorithms

Contact Info

Product

Resources

About