Sohan Lal

Sohan Lal is a postdoctoral researcher at the Technical University of Berlin (TU Berlin), focusing on advanced modeling and runtime support for large-scale HPC clusters under a DFG-funded project. His PhD in Computer Engineering from TU Berlin (2019) explored power modeling and architectural techniques for energy-efficient GPUs. He contributed to EU-funded LPGPU projects on low-power GPU computing, leading tasks and collaborating across consortium members. Previously, he lectured at Shri Mata Vaishno Devi University and worked as an IT specialist in the Government of India.

Education:

• PhD in Computer Engineering, TU Berlin (2019)
• Masters in Computer Science, IIT Delhi (2011)
• Bachelor in Computer Science and Engineering, GCET Jammu (2003)

His research interests span GPU architecture, power/performance modeling, memory systems, and applied machine learning. Notable contributions include techniques like Selective Lossy Compression (SLC) for GPUs and entropy encoding-based memory compression (E²MC). He received HiPEAC travel/grants and was an ACM SRC semifinalist (2018).

Grants & Collaborations:

• HiPEAC Collaboration Grant for joint work with TU/e
• DFG-funded postdoctoral research

He actively teaches advanced computer architectures and multicore systems at TU Berlin, reflecting his passion for education developed during his early teaching career.