Beyond Throughput: Performance and Energy Insights of LLM Inference Across AI Accelerators

Giacomo Brunetta, Varuni Sastry 0001, Xingfu Wu, Valerie Taylor 0001, Michael E. Papka, Zhiling Lan. Beyond Throughput: Performance and Energy Insights of LLM Inference Across AI Accelerators. In IEEE International Parallel and Distributed Processing Symposium, IPDPS 2026, New Orleans, LA, USA, May 25-29, 2026. pages 970-982, IEEE, 2026. [doi]

Authors

Giacomo Brunetta

This author has not been identified. Look up 'Giacomo Brunetta' in Google

Varuni Sastry 0001

This author has not been identified. Look up 'Varuni Sastry 0001' in Google

Xingfu Wu

This author has not been identified. Look up 'Xingfu Wu' in Google

Valerie Taylor 0001

This author has not been identified. Look up 'Valerie Taylor 0001' in Google

Michael E. Papka

This author has not been identified. Look up 'Michael E. Papka' in Google

Zhiling Lan

This author has not been identified. Look up 'Zhiling Lan' in Google