Skip to main navigation Skip to search Skip to main content

Multi-Dimensional ML-Pipeline Optimization in Cost-Effective Disaggregated Datacenter

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Machine learning (ML) pipelines deployed in datacenters are becoming increasingly complex and resource intensive, requiring careful optimizations to meet performance and latency requirements. Deployment in NUMA architectures with heterogeneous memory types such as CXL introduces a vast configuration space for memory and thread management. However, existing methods often need help to efficiently navigate this space, relying on significant manual tuning and needing more adaptability across diverse hardware platforms and extensive user-side code modification. This paper presents and experimentally evaluates an adaptive auto-tuning framework that optimizes memory configurations to maximize system throughput while adhering to Service Level Agreements (SLAs) on latency. Our optimization is carried out in two phases, addressing both performance and power efficiency. The proposed framework integrates an extended Berkeley Packet Filter (eBPF)-based kernel module for real-time performance monitoring and a user-space optimization core leveraging Bayesian Optimization and Pareto Optimality to explore high-dimensional configuration spaces. Extensive experimental evaluations on various workloads demonstrate that our framework achieves up to a 48% increase in throughput compared to default NUMA, TPP, Caption, and GPU while reducing search costs by as much as 77%. Furthermore, our two-phase optimization approach incorporates power efficiency, achieving up to 14.3% power savings without compromising throughput.

Original languageEnglish (US)
Title of host publicationMICRO 2025 - 58th IEEE/ACM International Symposium on Microarchitecture
PublisherIEEE Computer Society
Pages1178-1192
Number of pages15
ISBN (Electronic)9798400715730
DOIs
StatePublished - Oct 17 2025
Event58th IEEE/ACM International Symposium on Microarchitecture , MICRO 2025 - Seoul, Korea, Republic of
Duration: Oct 18 2025 → Oct 22 2025

Publication series

NameProceedings of the Annual International Symposium on Microarchitecture, MICRO
VolumePart of 213862
ISSN (Print)1072-4451

Conference

Conference58th IEEE/ACM International Symposium on Microarchitecture , MICRO 2025
Country/TerritoryKorea, Republic of
CitySeoul
Period10/18/25 → 10/22/25

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture

Fingerprint

Dive into the research topics of 'Multi-Dimensional ML-Pipeline Optimization in Cost-Effective Disaggregated Datacenter'. Together they form a unique fingerprint.

Cite this