Green: a framework for supporting energy-conscious programming using controlled approximation

By Woongki Baek, Trishul M. Chilimbi. ACM-SIGPLAN Symposium on Programming Language Design and Implementation, 2010

493 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Hardware and Architecture, Information Systems

View Paper: https://www.semanticscholar.org/paper/99b6bb4172d3332d231e4f2b2e288e2f305f0135

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Hardware and Architecture

In Information Systems

Citations in the corpus

EnerJ: approximate data types for safe and general low-power computation (2011, methodology): “Relatedly, EnerJ’s support for algorithmic approximation, the ability to write an approximate implementation and a precise implementation of the same functionality, bears some similarity to the Green programming model [3].”
Approximate Computing: A Survey (2016, methodology): “SAGE [11] and GREEN [22] check the output quality once in every N invocations of the approximate kernel, in which they compare the quality between approximate computation and exact computation and adjust the approximate modes for the subsequent computations accordingly.”
Analysis and characterization of inherent application resilience for approximate computing (2013, methodology): “Approximate computing techniques in software [10, 11, 12] have been proposed to improve the performance and parallel scalability of applications on general-purpose computing platforms.”
A Survey of Techniques for Approximate Computing (2016, methodology): “Baek and Chilimbi [2010] present a programming model framework that approximates functions and loops in a program and provides statistical guarantees of meeting the user-speciﬁed QoS target.”
Neural Acceleration for General-Purpose Approximate Programs (2012, unknown): “float 3 x = (p[0][0] + 2 ∗ p[0][1] + p[0][2]); x += (p[2][0] + 2 ∗ p[2][1] + p[2][2]); 5 y = (p[0][2] + 2 ∗ p[1][2] + p[2][2]); y += (p[0][0] + 2 ∗ p[1][1] + p[2][0]); 7 r = sqrt (x ∗ x + y ∗ y);”
Flikker: saving DRAM refresh-power through critical data partitioning (2011, background): “Flikker can also be applied to data-center applications because they, (1) exhibit high variations in workloads and have considerable periods of inactivity, (2) consume significant power in idle-mode due to over-engineering, and (3) are inherently error-resilient and do not have to be 100% accurate [”
Post-compiler software optimization for reducing energy (2014, background): “As mentioned earlier, there is increasing willingness to trade guarantees of exact semantic preservation for other desired system properties [1, 2, 5, 6, 8, 39, 40, 50, 58, 59].”