this paper we outline our framework for managing the dynamic precision variation. We represent the variation in the required precision for an operation by using a precision variation curve. The precision variation curve quantifies the variation in the required precision for an operation over time. The concept of time can represented by using various measures such as execution time, program counter, loop counter, etc. In this paper we analyze the variation of precision in loop computations as the iterations of the loops progress. Compile-time and run-time techniques to determine the precision variation curve for a given computation are described. Various algorithmic techniques are developed for optimal mapping of the computations onto reconf...
A new operation model of flexible calculation that allows us to adjust the operation delay depending...
This paper presents an e cient hardware algorithm for variable-precision division. The algorithm is ...
Summarization: The most popular representative devices of reconfigurable computing are field-program...
Reconfigurable architectures promise significant performance benefits by customizing the configurati...
Using standard Floating-Point (FP) formats for computation leads to significant hardware overhead si...
Customisable data formats provide an opportunity for exploring trade-offs in accuracy and performanc...
Abstract — In this paper, the authors discuss fine-grain power savings obtained through a technique ...
The precision used in an algorithm affects the error and performance of individual computations, the...
Dynamic hardware generation reduces the number of FPGA resources needed and speeds up an application...
International audienceFloating-Point (FP) computation using standard IEEE formats has a significant ...
. Reconfigurable circuits and systems have evolved from application specific accelerators to a gener...
Abstract-Currently, few tools exist to aid the FPGA developer in translating an algorithm designed f...
Modern communication systems such as 5G need high computational accuracy and dynamic range. Floating...
Many classes of applications, both in the embedded and high performance domains, can trade off the a...
Mixed precision is an approximate computing technique that can be used to trade-off computation accu...
A new operation model of flexible calculation that allows us to adjust the operation delay depending...
This paper presents an e cient hardware algorithm for variable-precision division. The algorithm is ...
Summarization: The most popular representative devices of reconfigurable computing are field-program...
Reconfigurable architectures promise significant performance benefits by customizing the configurati...
Using standard Floating-Point (FP) formats for computation leads to significant hardware overhead si...
Customisable data formats provide an opportunity for exploring trade-offs in accuracy and performanc...
Abstract — In this paper, the authors discuss fine-grain power savings obtained through a technique ...
The precision used in an algorithm affects the error and performance of individual computations, the...
Dynamic hardware generation reduces the number of FPGA resources needed and speeds up an application...
International audienceFloating-Point (FP) computation using standard IEEE formats has a significant ...
. Reconfigurable circuits and systems have evolved from application specific accelerators to a gener...
Abstract-Currently, few tools exist to aid the FPGA developer in translating an algorithm designed f...
Modern communication systems such as 5G need high computational accuracy and dynamic range. Floating...
Many classes of applications, both in the embedded and high performance domains, can trade off the a...
Mixed precision is an approximate computing technique that can be used to trade-off computation accu...
A new operation model of flexible calculation that allows us to adjust the operation delay depending...
This paper presents an e cient hardware algorithm for variable-precision division. The algorithm is ...
Summarization: The most popular representative devices of reconfigurable computing are field-program...