Metraipolis › archive › dc_49de3009d6ca

DVFS for Small Language Model Inference on Mobile Edge Devices — arXiv:2609.13153

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.13153
License (as archived)
CC0-1.0
Archived
2026-10-15T23:17:42.624Z
SHA-256 of the archived text
7783d13d19227245f346d66b64e0dbdc5ac54542cc3a5e7eb927b4fc07d7254a
Stable URL of this page
https://metraipolis.com/documento/dc_49de3009d6ca

How to cite

DVFS for Small Language Model Inference on Mobile Edge Devices — arXiv:2609.13153 — arXiv (en), https://arxiv.org/abs/2609.13153 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_49de3009d6ca (Metraipolis archive dc_49de3009d6ca, sha256 7783d13d19227245f346d66b64e0dbdc5ac54542cc3a5e7eb927b4fc07d7254a)
BibTeX
@misc{dc49de3009d6ca,
  title = {DVFS for Small Language Model Inference on Mobile Edge Devices — arXiv:2609.13153},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.13153},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_49de3009d6ca; sha256 7783d13d19227245f346d66b64e0dbdc5ac54542cc3a5e7eb927b4fc07d7254a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.13153. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

This paper presents DVFSLM, a new dynamic voltage and frequency scaling (DVFS) design for energy-efficient inference of small language models (SLMs) on mobile edge devices. The growing demand for local execution of language models has driven the adoption of SLMs, which balance computational feasibility with good inference performance. However, energy efficiency remains a critical challenge, since even miniaturized SLMs impose significant energy consumption, impacting application quality, device reliability, and environmental sustainability. Existing DVFS solutions, designed for cloud-based large models or generic mobile workloads, fail to address the unique workload characteristics of SLMs, resulting in wasted energy or excessive latency. Unlike prior work, DVFSLM explicitly addresses two key challenges: 1) the complex interdependencies of processor frequencies, power and latency across autoregressive token generations, and 2) hardware opacity, where the individual power and latency contributions from different processors (GPU, CPU and EMC) are obscured during collaborative execution. To address these, DVFSLM introduces workload-aware power and latency estimators that analyze core matrix operations and correlate them with hardware metadata, enabling precise estimations of how frequency adjustments impact power and latency. These estimations drive a runtime DVFS governor that coordinates the GPU and EMC frequencies with a profiled CPU-frequency threshold, minimizing the energy per token while satisfying configurable token-generation deadlines. Extensive experiments on a rich set of SLMs show that DVFSLM reduces the energy per token by up to 12.4% over the latest built-in governors and up to 8.4% over the state-of-the-art GearDVFS, while improving the latency quality of service (QoS) by up to 93.12% and 69.14%, respectively.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-09-15 in cs.DC).
Categorie arXiv: Hardware Architecture (cs.AR) ; Distributed, Parallel, and Cluster Computing (cs.DC).
Pagina del lavoro: https://arxiv.org/abs/2609.13153 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-15T22:54:12.998Z dalla pagina https://arxiv.org/catchup/cs.DC/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).