Computing systems have made an irreversible transition towards parallel architectures with the multi-cores and many cores.
However, power and thermal limits are rapidly bringing the computing community to another crossroad where a chip can have
many cores but a significant fraction of them are left unpowered --- or dark --- at any point in time. This phenomenon,
known as dark silicon, is driving the emergence of heterogeneous platforms consisting of cores with diverse power-performance
characteristics enabling better match between application requirements and computation capabilities leading to substantially
improved energy-efficiency.
This project aim to design a polymorphic heterogeneous multi-core architecture, called Bahurupi, which can be tailored according to
the application by software. Bahurupi is designed and fabricated as a homogeneous multi-core system containing multiple identical
(simple) cores on chip along with some amount of reconfigurable logic. The main novelty of Bahurupi lies in its highly flexible
architecture. Post-fabrication, software can configure or compose together primitive cores to create a customized multi-core system that best matches the
needs of a specific application. Moreover, the functionality of the cores can be extended, again at runtime, through application-specific custom instructions
implemented in the reconfigurable fabric on-chip. This software directed re-configurability allows us to enjoy the power-performance benefits of a
customized and optimized heterogeneous multi-core system without paying the hefty price of design and fabrication for customization.
[DAC] Hierarchical Power Management for Asymmetric Multi-Core in Dark Silicon Era
Thannirmalai Somu Muthukaruppan, Mihai Pricopi, V Vanchinathan, Tulika Mitra, Sanjay Vishin
50th Design Automation Conference, June 2013
[FPT]
Graph Minor Approach for Application Mapping on CGRAs
Liang Chen, Tulika Mitra
International Conference on Field Programmable Technology, December 2012
Best Paper Award
[TACO]
Bahurupi: A polymorphic heterogeneous multi-core architecture
Mihai Pricopi, Tulika Mitra
ACM Transactions on Architecture and Code Optimization, 8(4), January 2012
Presented at 7th International Conference on High-Performance and Embedded Architectures and Compilers (HiPEAC) 2012
Best Student Poster Award
[DAC]
Shared Reconfigurable Fabric for Multi-Core Customization<
Liang Chen, Tulika Mitra
Design Automation Conference, June 2011
An Efficient Framework for Dynamic Reconfiguration of Instruction-Set Customization
Huynh Phung Huynh, Edward Sim, Tulika Mitra
Design Automation for Embedded Systems, Springer Journal, 13(1-2), June 2009
Invited: Special Issue on Best Papers from ESWeek 2007
[FPL] Disjoint Pattern Enumeration for Custom Instructions Identification [Technical report]
Pan Yu, Tulika Mitra
International Conference on Field Programmable Logic and Applications, August 2007
Best Paper Award Nomination
[CASES]
Scalable Custom Instructions Identification for Instruction-Set Extensible Processors,
Pan Yu, Tulika Mitra
International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, September 2004
[DAC]
Characterizing Embedded Applications for Instruction-Set Extensible Processors
Pan Yu, Tulika Mitra
Design Automation Conference, June 2004