Litcius/Paper detail

TAPAS: Thermal- and Power-Aware Scheduling for LLM Inference in Cloud Platforms

Jovan Stojkovic, Chaojie Zhang, Íñigo Goiri, Esha Choukse, Haoran Qiu, Rodrigo Fonseca, Josep Torrellas, Ricardo Bianchini

202520 citationsDOIOpen Access PDF

Abstract

The rising demand for generative large language models (LLMs) poses challenges for thermal and power management in cloud datacenters. Traditional techniques are often inadequate for LLM inference due to the fine-grained, millisecond-scale execution phases, each with distinct performance, thermal, and power profiles. Additionally, LLM inference workloads are sensitive to various configuration parameters (e.g., model parallelism, size, and quantization) that involve trade-offs between performance, temperature, power, and output quality. Moreover, clouds often co-locate SaaS and IaaS workloads, each with different levels of visibility and flexibility.

Topics & Concepts

Computer scienceCloud computingInferenceScheduling (production processes)Distributed computingOperating systemArtificial intelligenceEngineeringOperations managementCloud Computing and Resource ManagementAdvanced Data Storage TechnologiesSoftware System Performance and Reliability
TAPAS: Thermal- and Power-Aware Scheduling for LLM Inference in Cloud Platforms | Litcius