Metraipolis › archive › dc_493425bea63a

CARB: A Characterization-Guided Framework for CNN Inference Cost Prediction and Deployment Screening — arXiv:2608.10506

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.10506
License (as archived)
CC0-1.0
Archived
2026-10-19T06:14:39.852Z
SHA-256 of the archived text
0e9c6a7fdd5ac273a4b0fbc792efeadaaee1c81fff6564e6d8203025fb442f9e
Stable URL of this page
https://metraipolis.com/documento/dc_493425bea63a

How to cite

CARB: A Characterization-Guided Framework for CNN Inference Cost Prediction and Deployment Screening — arXiv:2608.10506 — arXiv (en), https://arxiv.org/abs/2608.10506 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_493425bea63a (Metraipolis archive dc_493425bea63a, sha256 0e9c6a7fdd5ac273a4b0fbc792efeadaaee1c81fff6564e6d8203025fb442f9e)
BibTeX
@misc{dc493425bea63a,
  title = {CARB: A Characterization-Guided Framework for CNN Inference Cost Prediction and Deployment Screening — arXiv:2608.10506},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.10506},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_493425bea63a; sha256 0e9c6a7fdd5ac273a4b0fbc792efeadaaee1c81fff6564e6d8203025fb442f9e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.10506. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Accurate pre-deployment estimation of CNN inference cost--energy, latency, and peak memory--is increasingly critical as models are deployed on resource-constrained GPU platforms. Existing approaches rely on FLOPs, latency measurements, or single-device profiling as energy proxies, overlooking the non-linear interactions between architectural design and hardware load. We present a workload characterization study of 13 419 CNN configurations on two GPU platforms (RTX 5090 and RTX 3080) under GPU telemetry, revealing that energy, latency, and memory exhibit fundamentally distinct scaling behaviors: energy and latency diverge by 3x under high computational demand, and cross-GPU transferability differs by target--energy and latency require platform-specific models while memory transfers well across the two tested platforms. Building on these characterization findings, we develop CARB, a cascade-blended ensemble that jointly predicts all three targets with R2 ~0.99, and a two-stage deployment screening workflow that eliminates over 90% of candidates in seconds, reducing large design spaces to a Pareto-prioritized shortlist validated against real hardware.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-08-12 in cs.AR).
Categorie arXiv: Hardware Architecture (cs.AR) ; Machine Learning (cs.LG); Performance (cs.PF).
Pagina del lavoro: https://arxiv.org/abs/2608.10506 — 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 2026-10-19T06:09:22.678Z dalla pagina https://arxiv.org/catchup/cs.AR/2026-08-12?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).