Kiran Gunnam

According to our database, Kiran Gunnam authored at least 3 papers in 2025.

Collaborative distances:

• Dijkstra number of six.
• Erdős number of six.

2025

LLM Training Workload IO Characteristics.

Kiran Gunnam, Alex Mohandas, Mahesh Kumar Dhote, Rajesh Bhagwat, Darshan P, Kanaiah Kothalikar

Proceedings of the IEEE International Symposium on Circuits and Systems, 2025

Enhancing Training Efficiency: A Novel Approach to Handling GPU Failures in Large-Scale Distributed System for LLM Training.

Kiran Gunnam, Pranshu Tapan Mandal, Shivam Khandelwal, Rajesh Bhagwat

Proceedings of the 7th IEEE International Conference on Artificial Intelligence Circuits and Systems, 2025

Addition of CXL memory pool to increase LLM training check-pointing frequency.

Kiran Gunnam, Prasad Chandrashekara, Murali Kothandapani, Rajesh Bhagwat, Priyanshi Jain

Proceedings of the 7th IEEE International Conference on Artificial Intelligence Circuits and Systems, 2025