We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.IT

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Information Theory

Title: Towards Efficient Large-Scale Network Slicing: An LP Rounding-and-Refinement Approach

Abstract: In this paper, we propose an efficient algorithm for the network slicing problem which attempts to map multiple customized virtual network requests (also called services) to a common shared network infrastructure and allocate network resources to meet diverse service requirements. The problem has been formulated as a mixed integer linear programming (MILP) formulation in the literature. We first propose a novel linear programming (LP) relaxation of the MILP formulation. We show that compared with a natural LP relaxation of the MILP formulation, the novel LP relaxation is much more compact in terms of smaller numbers of variables and constraints, and much stronger in terms of providing a better LP bound, which makes it particularly suitable to be embedded in an LP based algorithm. Then we design an efficient two-stage LP rounding-and-refinement algorithm based on this novel LP relaxation. In the first stage, the proposed algorithm uses an iterative LP rounding procedure to place the virtual network functions of all services into cloud nodes while taking traffic routing of all services into consideration; in the second stage, the proposed algorithm uses an iterative LP refinement procedure to obtain a solution for traffic routing of all services with their end-to-end delay constraints being satisfied. Compared with the existing algorithms which either have an exponential complexity or return a low-quality solution, our proposed algorithm achieves a better trade-off between the solution quality and the computational complexity. In particular, the worst-case complexity of our proposed algorithm is polynomial, which makes it suitable for solving large-scale problems. Numerical results demonstrate the effectiveness and efficiency of our proposed algorithm.
Comments: 13 pages, 9 figures. arXiv admin note: substantial text overlap with arXiv:2102.02563
Subjects: Information Theory (cs.IT); Signal Processing (eess.SP); Optimization and Control (math.OC)
Cite as: arXiv:2107.14404 [cs.IT]
  (or arXiv:2107.14404v2 [cs.IT] for this version)

Submission history

From: Wei-Kun Chen [view email]
[v1] Fri, 30 Jul 2021 02:50:48 GMT (55kb)
[v2] Sun, 27 Feb 2022 07:54:40 GMT (67kb)

Link back to: arXiv, form interface, contact.