Publications

Under Submission: Titles changed to maintain anonymity


  • Control Flow Structuring without Code Blowup

-----------------------------------------------------------------------

Peer reviewed conferences/journals/workshops


Fast Distributed Bandits for Online Recommendations [pdf]

Kanak Mahadik Qingyun Wu, Shuai Li, Amit Sabne

ACM Internationcal Conference of Supercomputing (ICS), 2020


  • Pagoda: A GPU Runtime System for Narrow Tasks [pdf]

Tsung Tai Yeh, Amit Sabne, Putt Sakdhnagool, Rudolf Eigenmann, and Timothy Rogers

ACM Transactions on Parallel Processing (TOPC), 2019


  • Logic Synthesis of Approximate Circuits [pdf]

Swagath Venkataramani, Vivek Joy, Amit Sabne, Kaushik Roy, and Anand Raghunathan

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2019


  • Comparative Analysis of Coprocessors [pdf]

Putt Sakdhnagool, Amit Sabne, and Rudolf Eigenmann

Concurrency and Computation: Practice and Experience, 2018


  • Massively Parallel 3D Image Reconstruction [pdf] [ppt] [news]

Xiao Wang, Amit Sabne, Putt Sakdhnagool, Sherman Kisner, Charles Bowman, and Samuel Midkiff

ACM/IEEEInternational Conference for High Performance Computing, Networking, Storage and Analysis (SC), 2017, Gordon Bell Paper Award Finalist

  • Model-based Iterative CT Image Reconstruction on GPUs [pdf] [ppt]

Amit Sabne, Xiao Wang, Sherman Kisner, Anand Raghunathan, Charles Bowman, and Samuel Midkiff

ACM Symposium on Principles and Practice of Parallel Programming (PPoPP), 2017


  • Pagoda: Fine-grained GPU Resource Virtualization for Narrow Tasks [pdf] [ppt]

Tsung Tai Yeh, Amit Sabne, Putt Sakdhnagool, Rudolf Eigenmann, and Timothy Rogers

ACM Symposium on Principles and Practice of Parallel Programming (PPoPP), 2017, Best Paper Award Nominee


  • Formalizing Structured Control Flow Graphs [pdf] [ppt]

Amit Sabne, Putt Sakdhnagool, and Rudolf Eigenmann

International Workshop on Languages and Compilers for Parallel Computing (LCPC), 2016


  • High Performance Model Based Reconstruction [pdf]

Xiao Wang, Amit Sabne, Sherman Kisner, Anand Raghunathan, Charles Bowman, and Samuel Midkiff

ACM Symposium on Principles and Practice of Parallel Programming (PPoPP), 2016


  • Hydra : Extending Shared Address Programming for Accelerator Clusters [ppt] [pdf]

Putt Sakdhnagool, Amit Sabne, and Rudolf Eigenmann

International Workshop on Languages and Compilers for Parallel Computing (LCPC), 2015


  • Understanding Portability of a High-level Programming Model on Contemporary Heterogeneous Architectures [pdf]

Amit Sabne, Putt Sakdhnagool, Seyong Lee, and Jeffrey S. Vetter

IEEE Micro Magazine, July-August, 2015


  • HeteroDoop: A MapReduce Programming System for Accelerator Clusters [ppt] [pdf]

Amit Sabne, Putt Sakdhnagool, and Rudolf Eigenmann

ACM International Symposium on High-Performance Parallel and Distributed Computing (HPDC), 2015


  • Evaluating Performance Portability of OpenACC [ppt] [pdf]

Amit Sabne, Putt Sakdhnagool, Seyong Lee, and Jeffrey S. Vetter

International Workshop on Languages and Compilers for Parallel Computing (LCPC), 2014


  • Scaling Large-Data Computations on Multi-GPU Accelerators [ppt] [pdf]

Amit Sabne, Putt Sakdhnagool, and Rudolf Eigenmann

ACM International Conference on Supercomputing (ICS), 2013


  • Effects of Compiler Optimizations in OpenMP to CUDA Translation [ppt] [pdf]

Amit Sabne, Putt Sakdhnagool, and Rudolf Eigenmann

International Workshop on OpenMP (IWOMP), 2012


  • SALSA: Systematic Logic Synthesis of Approximate Circuits [pdf]

Swagath Venkataramani, Amit Sabne, Vivek Joy, Kaushik Roy, and Anand Raghunathan

Design and Automation Conference (DAC), 2012


  • A Generic Low Power Scan Chain Wrapper for Designs Using Scan Compression [pdf]

Amit Sabne, Rajesh Tiwari, Abhijeet Shrivastava, Srivaths Ravi, and Rubin Parekhji

IEEE VLSI Test Symposium (VTS), 2010


  • Embedded Systems Applications in Robotics and Automation [pdf]

Amit Sabne and Amalin A Prince

National Conference on Emerging Trends and Developments in Embedded Systems, Goa: India, 2007 Best Paper Award


Posters:

  • Optimizing GPU Programs by Register Demotion [pdf]

Putt Sakdhnagool, Amit Sabne, and Rudolf Eigenmann

ACM Symposium of Principles and Practice of Parallel Programming (PPoPP), 2019


  • Pagoda: A Runtime System to Maximize Utilization in Data Parallel Tasks with Limited Parallelism [pdf]

Tsung Tai Yeh, Amit Sabne, Putt Sakdhnagool, Rudolf Eigenmann, and Timothy Rogers

International Conference on Parallel Architectures and Compilation (PACT), 2016


  • Hydra: Extending Shared Address Programming for Accelerator Clusters [pdf]

Putt Sakdhnagool, Amit Sabne, and Rudolf Eigenmann

International Workshop on Languages and Compilers for Parallel Computing (LCPC), 2015


  • HeteroDoop: Automatically Exploiting CPUs and GPUs for Big Data MapReduce [pdf]

Amit Sabne and Rudolf Eigenmann

GPU Technology Conference (GTC), 2014


Other articles:

  • RegDem: Increasing GPU Performance by Shared Memory Register Spilling [arxiv]

Putt Sakdhnagool, Amit Sabne, and Rudolf Eigenmann