Abstract
Prior work demonstrated the use of specialized processors, or accelerators, be energy-efficient for binary floatingpoint (BFP) division and square root, and for decimal floatingpoint (DFP) operations. In the dark silicon era, where not all the circuits on the die can be powered simultaneously, we propose a hybrid BFP/DFP engine to perform BFP division and DFP addition, multiplication and division. The main purpose of this engine is to offload the binary floating-point units for this type of operations and reduce the latency for decimal operations, and power and temperature for the whole die.
Original language | English |
---|---|
Title of host publication | Proceedings of the 22nd IFIP/IEEE International Conference on Very Large Scale Integration (VLSI-SoC) |
Number of pages | 6 |
Publisher | IEEE |
Publication date | 2014 |
ISBN (Print) | 978-1-4799-6016-3 |
DOIs | |
Publication status | Published - 2014 |
Event | 22nd IFIP/IEEE International Conference on Very Large Scale Integration - Playa del Carmen, Mexico Duration: 6 Oct 2014 → 8 Oct 2014 Conference number: 22 http://xilonen.inaoep.mx/vlsi-soc2014/ |
Conference
Conference | 22nd IFIP/IEEE International Conference on Very Large Scale Integration |
---|---|
Number | 22 |
Country/Territory | Mexico |
City | Playa del Carmen |
Period | 06/10/2014 → 08/10/2014 |
Internet address |