Publications [Google Scholar]

2024

A Portable Compiler-Runtime Approach for Scalability Prediction (Under Review at TACO)

Kowshic Ahmed Akash, Tobias Wulf, Torsten Valentin, Alexander Geist, Ulf Kulau, Sohan Lal
Pixels to Precision: Harnessing AI for Anomaly Detection in Post-Silicon Validation Accepted as work in progress (WIP) at the International Design Automation Conference (DAC) 2024

Tim Lühnen, Alexander Brosig, Sohan Lal
Auto-TBC: Auto-Optimizing CUDA Applications Using Thread Block Clusters Accepted for presentation as research poster at the International Supercomputing Conference (ISC) 2024

2023

Manuel Renz, Sohan Lal
Beyond Compression Ratio: A Throughput Analysis of Memory Compression Techniques for GPUs Accepted for Publication at the Proceedings of International Conference on Computer Design (ICCD’23), Washington DC, USA

Apurv Ashish, Sohan Lal, Ben Juurlink
An Efficient Lightweight Framework for Porting Vision Algorithms on Embedded SoCs
International Embedded Systems Symposium (IESS), Germany

2022

Sohan Lal, Bogaraju Sharatchandra Varma, and Ben Juurlink
A Quantitative Study of Locality in GPU Caches for Memory-Divergent Workloads
Published at the International Journal of Parallel Programming (IJPP)

Sohan Lal, Manuel Renz, Julian Hartmer, and Ben Juurlink
Memory Access Granularity Aware Lossless Compression for GPUs
In the Proceedings of International Conference on Parallel and Distributed Processing Symposium (IPDPS’22), France

2021

Sohan Lal, Jan Lucas, Ben Juurlink
QSLC: Quantization-Based, Low-Error Selective Approximation for GPUs
In the Proceedings of Design, Automation & Test in Europe Conference & Exhibition (DATE 2021), Virtual Event

2020

Sohan Lal, Aksel Alpay, Philip Salzmann, Biagio Cosenza, Nicolai Stawinoga, Peter Thoman, Thomas Fahringer and Vincent Heuveline
SYCL-Bench: A Versatile Cross-Platform Benchmark Suite for Heterogeneous Computing
In the Proceedings of International European Conference on Parallel and Distributed Computing (Euro-Par’20), Virtual Event

Sohan Lal, Ben Juurlink
A Quantitative Study of Locality in GPU Caches
In the Proceedings of International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation (SAMOS’20), Virtual event

Sohan Lal, Aksel Alpay, Philip Salzmann, Biagio Cosenza, Nicolai Stawinoga, Peter Thoman, Thomas Fahringer and Vincent Heuveline
SYCL-Bench: A Versatile Single-Source Benchmark Suite for Heterogeneous Computing
Extended Abstract at Workshop on OpenCL (IWOCL/SYCLcon’20), Germany

2019

Nadjib Mammeri, Markus Neu, Sohan Lal, Ben Juurlink
Performance Counters based Power Modeling of Mobile GPUs using Deep Learning
In the Proceedings of International Conference on High Performance Computing and Simulation (HPCS’19), Ireland

Sohan Lal, Jan Lucas, Ben Juurlink
SLC: Memory Access Granularity Aware Selective Lossy Compression for GPUs
In the Proceedings of International Conference on Design Automation, and Test in Europe (DATE’19), Italy

Sohan Lal
Power Modeling and Architectural Techniques for Energy-Efficient GPUs
PhD Thesis, Technische Universität Berlin, 2019

2018

Sohan Lal, Ben Juurlink
A Case for Memory Access Granularity Aware Selective Lossy Compression for GPUs
ACM Student Research Competition Poster and Extended Abstract at IEEE/ACM International Symposium on Microarchitecture (MICRO’18), Japan, (Semifinalist)

Jan Lucas, Sohan Lal, Ben Juurlink
Optimal DC/AC Data Bus Inversion Coding
In the Proceedings of International Conference on Design Automation, and Test in Europe (DATE’18), Germany

2017

Sohan Lal, Jan Lucas, Ben Juurlink
E2MC: Entropy Encoding Based Memory Compression for GPUs
In the Proceedings of International Conference on Parallel and Distributed Processing Symposium (IPDPS’17), USA

2016

Maurice Peemen, Runbin Shi, Sohan Lal, Ben Juurlink, Bart Mesman, and Henk Corporaal
The Neuro Vector Engine: Flexibility to Improve Convolutional Net Efficiency for Wearable Vision
In the Proceedings of International Conference on Design Automation, and Test in Europe (DATE’16), Germany

2014

Sohan Lal, Jan Lucas, Michael Andersch, Mauricio Alvarez Mesa, Ahmed Elhossini, and Ben Juurlink
GPGPU Workload Characteristics and Performance Analysis
In the Proceedings of International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation (SAMOS’14), Greece

2013

Jan Lucas, Sohan Lal, Michael Andersch, Mauricio Alvarez Mesa, and Ben Juurlink
How a Single Chip Causes Massive Power Bills - GPUSimPow: A GPGPU Power Simulator
In the Proceedings of International Symposium on Performance Analysis of Systems and Software (ISPASS’13), USA

Sohan Lal, Jan Lucas, Mauricio Alvarez Mesa, Ahmed Elhossini, and Ben Juurlink
Exploring GPGPUs Workload Characteristics and Power Consumption
International Summer School on Advanced Computer Architecture and Compilation for High-Performance and Embedded Systems (ACACES’13), Italy

Jan Lucas, Sohan Lal, Mauricio Alvarez Mesa, Ahmed Elhossini, and Ben Juurlink
DART: A GPU Architecture Exploiting Temporal SIMD for Divergent Workloads
International Summer School on Advanced Computer Architecture and Compilation for High-Performance and Embedded Systems (ACACES’13), Italy

Sohan Lal, Kolin Paul, James Gomes
Accelerating Metabolic Pathways Simulation using GPUs
In the International Conference on Advances in Distributed and Parallel Computing (ADPC’10), Singapore