Journal article · 2026 · Open access

A Provably Cost-Efficient Approach to Deploying MoE Inference Models at the Network Edge

Chao Wang, Danyang Zheng, Huanlai Xing, Chen Yang, Xiaojun Cao, Jie Xu, Fei Teng

IEEE Transactions on Services Computing, 19(4), 2646–2659 · Published 25 June 2026 · DOI 10.1109/tsc.2026.3707550

Open full textCite this paper

In brief

“A Provably Cost-Efficient Approach to Deploying MoE Inference Models at the Network Edge” is a 2026 journal article by Chao Wang and 6 co-authors, published in IEEE Transactions on Services Computing. OpenAlex counts 1 citations as of 19 November 2026. It is open access under a CC BY licence: the full text is free to read at IEEE Transactions on Services Computing.

Citations
1
Most cited in
2026 (1)
Field-weighted impact
1.68
References
48
Authors
7

Abstract

Foundation model inference is rapidly turning into a pay-per-request utility, yet today’s infrastructure still deploys experts as dedicated, always-on services. In Mixture-of-Experts models, sparse activation makes per-request execution paths highly dynamic, which amplifies provisioning uncertainty while creating opportunities for Inner-Service Resource Sharing (ISRS) across requests. This paper studies cost-efficient provisioning of MoE inference as a service over geo-distributed edge infrastructure, where a provider must satisfy the Service Level Agreement (SLA) constraints while minimizing the joint cost of compute, VRAM, and bandwidth. We formalize the Mixture-of-Experts Deployment (MoED) problem by explicitly modeling ISRS under dynamic routing and multicast communication. To solve MoED, we design an Expert Cost Loss (ECL) metric and a Sparsity-Aware Augmented Graph (SAG) that together guide ISRS-aware expert placement and bandwidth reservation, and we propose the ECL-based Expert Placement in SAG (ECL-EPS) algorithm. We prove that ECL-EPS achieves a logarithmic approximation ratio under practical assumptions. Extensive simulations show that ECL-EPS consistently reduces deployment cost and improves resource-sharing efficiency compared with the benchmarks.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Wang, C., Zheng, D., Xing, H., Yang, C., Cao, X., Xu, J., & Teng, F. (2026). A Provably Cost-Efficient Approach to Deploying MoE Inference Models at the Network Edge. IEEE Transactions on Services Computing, 19(4), 2646–2659. https://doi.org/10.1109/tsc.2026.3707550

MLA 9

Wang, Chao, et al. “A Provably Cost-Efficient Approach to Deploying MoE Inference Models at the Network Edge.” IEEE Transactions on Services Computing, vol. 19, no. 4, 2026, pp. 2646–2659. https://doi.org/10.1109/tsc.2026.3707550.

Chicago

Wang, Chao, Danyang Zheng, Huanlai Xing, Chen Yang, Xiaojun Cao, Jie Xu, and Fei Teng. 2026. “A Provably Cost-Efficient Approach to Deploying MoE Inference Models at the Network Edge.” IEEE Transactions on Services Computing 19 (4): 2646–2659. https://doi.org/10.1109/tsc.2026.3707550.

Harvard

Wang, C. et al. (2026) ‘A Provably Cost-Efficient Approach to Deploying MoE Inference Models at the Network Edge’, IEEE Transactions on Services Computing, 19(4), pp. 2646–2659. doi:10.1109/tsc.2026.3707550.

IEEE

C. Wang et al., “A Provably Cost-Efficient Approach to Deploying MoE Inference Models at the Network Edge,” IEEE Transactions on Services Computing, vol. 19, no. 4, pp. 2646–2659, 2026, doi: 10.1109/tsc.2026.3707550.

BibTeX

@article{wang2026provably,
  title     = {{A Provably Cost-Efficient Approach to Deploying MoE Inference Models at the Network Edge}},
  author    = {Wang, Chao and Zheng, Danyang and Xing, Huanlai and Yang, Chen and Cao, Xiaojun and Xu, Jie and Teng, Fei},
  journal   = {IEEE Transactions on Services Computing},
  year      = {2026},
  volume    = {19},
  number    = {4},
  pages     = {2646--2659},
  doi       = {10.1109/tsc.2026.3707550},
  url       = {https://doi.org/10.1109/tsc.2026.3707550}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing

  • mixture of experts
  • service level agreements
  • bandwidth reservation
  • sparse activation
  • approximation algorithms
  • multicast
  • dynamic routing

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Wang, C., Zheng, D., Xing, H., Yang, C., Cao, X., Xu, J., & Teng, F. (2026). A Provably Cost-Efficient Approach to Deploying MoE Inference Models at the Network Edge. IEEE Transactions on Services Computing, 19(4), 2646–2659. https://doi.org/10.1109/tsc.2026.3707550

How many times has this paper been cited?

1 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at IEEE Transactions on Services Computing: sign in and open the full text from this page.

Who wrote this paper?

Chao Wang, Danyang Zheng, Huanlai Xing, Chen Yang, Xiaojun Cao, Jie Xu, Fei Teng.

What is the DOI of this paper?

10.1109/tsc.2026.3707550, permanently at https://doi.org/10.1109/tsc.2026.3707550. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.