This paper introduces heuristic approaches and a deep reinforcement learning approach to solve a joint virtual network function deployment and scheduling problem in a dynamic scenario. We formulate the problem as an optimization problem. Based on the mathematical description of the optimization problem, we introduce three heuristic approaches and a deep reinforcement learning approach to solve the problem. We define an objective to maximize the ratio of delay-satisfied requests while minimizing the average resource cost for a dynamic scenario. Our introduced two greedy approaches are named finish time greedy and computational resource greedy, respectively. In the finish time greedy approach, we make each request be finished as soon as possible despite its resource cost; in the computational resource greedy approach, we make each request occupy as few resources as possible despite its finish time. Our introduced simulated annealing approach generates feasible solutions randomly and converges to an approximate solution. In our learning-based approach, neural networks are trained to make decisions. We use a simulated environment to evaluate the performances of our introduced approaches. Numerical results show that the introduced deep reinforcement learning approach has the best performance in terms of benefit in our examined cases.
Zixiao ZHANG
Kyoto University
Eiji OKI
Kyoto University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Zixiao ZHANG, Eiji OKI, "Joint Virtual Network Function Deployment and Scheduling via Heuristics and Deep Reinforcement Learning" in IEICE TRANSACTIONS on Communications,
vol. E106-B, no. 12, pp. 1424-1440, December 2023, doi: 10.1587/transcom.2023EBP3039.
Abstract: This paper introduces heuristic approaches and a deep reinforcement learning approach to solve a joint virtual network function deployment and scheduling problem in a dynamic scenario. We formulate the problem as an optimization problem. Based on the mathematical description of the optimization problem, we introduce three heuristic approaches and a deep reinforcement learning approach to solve the problem. We define an objective to maximize the ratio of delay-satisfied requests while minimizing the average resource cost for a dynamic scenario. Our introduced two greedy approaches are named finish time greedy and computational resource greedy, respectively. In the finish time greedy approach, we make each request be finished as soon as possible despite its resource cost; in the computational resource greedy approach, we make each request occupy as few resources as possible despite its finish time. Our introduced simulated annealing approach generates feasible solutions randomly and converges to an approximate solution. In our learning-based approach, neural networks are trained to make decisions. We use a simulated environment to evaluate the performances of our introduced approaches. Numerical results show that the introduced deep reinforcement learning approach has the best performance in terms of benefit in our examined cases.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.2023EBP3039/_p
Copy
@ARTICLE{e106-b_12_1424,
author={Zixiao ZHANG, Eiji OKI, },
journal={IEICE TRANSACTIONS on Communications},
title={Joint Virtual Network Function Deployment and Scheduling via Heuristics and Deep Reinforcement Learning},
year={2023},
volume={E106-B},
number={12},
pages={1424-1440},
abstract={This paper introduces heuristic approaches and a deep reinforcement learning approach to solve a joint virtual network function deployment and scheduling problem in a dynamic scenario. We formulate the problem as an optimization problem. Based on the mathematical description of the optimization problem, we introduce three heuristic approaches and a deep reinforcement learning approach to solve the problem. We define an objective to maximize the ratio of delay-satisfied requests while minimizing the average resource cost for a dynamic scenario. Our introduced two greedy approaches are named finish time greedy and computational resource greedy, respectively. In the finish time greedy approach, we make each request be finished as soon as possible despite its resource cost; in the computational resource greedy approach, we make each request occupy as few resources as possible despite its finish time. Our introduced simulated annealing approach generates feasible solutions randomly and converges to an approximate solution. In our learning-based approach, neural networks are trained to make decisions. We use a simulated environment to evaluate the performances of our introduced approaches. Numerical results show that the introduced deep reinforcement learning approach has the best performance in terms of benefit in our examined cases.},
keywords={},
doi={10.1587/transcom.2023EBP3039},
ISSN={1745-1345},
month={December},}
Copy
TY - JOUR
TI - Joint Virtual Network Function Deployment and Scheduling via Heuristics and Deep Reinforcement Learning
T2 - IEICE TRANSACTIONS on Communications
SP - 1424
EP - 1440
AU - Zixiao ZHANG
AU - Eiji OKI
PY - 2023
DO - 10.1587/transcom.2023EBP3039
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E106-B
IS - 12
JA - IEICE TRANSACTIONS on Communications
Y1 - December 2023
AB - This paper introduces heuristic approaches and a deep reinforcement learning approach to solve a joint virtual network function deployment and scheduling problem in a dynamic scenario. We formulate the problem as an optimization problem. Based on the mathematical description of the optimization problem, we introduce three heuristic approaches and a deep reinforcement learning approach to solve the problem. We define an objective to maximize the ratio of delay-satisfied requests while minimizing the average resource cost for a dynamic scenario. Our introduced two greedy approaches are named finish time greedy and computational resource greedy, respectively. In the finish time greedy approach, we make each request be finished as soon as possible despite its resource cost; in the computational resource greedy approach, we make each request occupy as few resources as possible despite its finish time. Our introduced simulated annealing approach generates feasible solutions randomly and converges to an approximate solution. In our learning-based approach, neural networks are trained to make decisions. We use a simulated environment to evaluate the performances of our introduced approaches. Numerical results show that the introduced deep reinforcement learning approach has the best performance in terms of benefit in our examined cases.
ER -