Download presentation
Presentation is loading. Please wait.
1
Green: A Framework for Supporting Energy-Conscious Programming using Controlled Approximation Woongki Baek Stanford University Trishul M. Chilimbi Microsoft Research PLDI 2010
2
General Problem Tradeoff between Quality of Service (QoS) and Performance + Energy Consumption Why is it critical? Datacenters: Amazon, Google, Microsoft Existence of acceptable domains: Machine Learning, Image/Video Processing Programmers do this anyway, but in an ad-hoc manner and no QoS guarantees Green Framework was built to address these issues
3
Green Framework Overview Static and dynamic calibration QoS Model Language Extension
4
Functionality Input provided by programmer: QoS loss (maximum acceptable) i.e. 2% QoS computation function through language extension Optionally: approximate version(s) of a function or a loop calibration mechanism Output by Green system: New version of a program that is tuned to be more efficient (both performance and power) QoS guarantees to be in the acceptable range Dynamic re-calibration (adaptation) if needed at run- time
5
Overview General Problem and Functionality System Design Green Implementation Benchmarks Evaluation Discussion
6
Green Framework Design QoS Service Level Agreement (SLA) QoS_Compute QoS_Approx and QoS-ReCalibrate
7
Green Mechanisms
8
Loop Approximation To be replaced
9
QoS_Compute Used in the Calibration Phase
10
QoS_Lp_Approx
11
QoS Model for Loops Calibration data goes to MATLAB program Automatically selects appropriate approximation level Provides 2 interfaces:
12
Function Approximation
13
Approximation Modeling For each function and loop individually Then extensive search space exploration to combine them and still fulfill QoS SLA Global recalibration based on QoS_Loss/Performance_Gain Exponential backoff scheme to avoid non- linear effects if any
14
Experiments Benchmarks: Bing Search Graphics: 252.eon (SPEC2000) Machine Learning: Cluster GA Signal Processing: Discrete Fourier Transform Finance: Blackscholes (PARSEC)
15
Results with Bing Search - Performance
16
Results with Bing Search - QoS
17
Results with 252.eon - Performance
18
Results with 252.eon - QoS
19
Results with 252.eon – Model Sensitivity
20
Results with DFT – Performance And Energy
21
Results with DFT – QoS
22
Discussion Advantages: Performance and power efficiency improvements Design flexibility Automatic QoS modeling Fine granularity: function and loop based
23
Discussion Disadvantages: Limited scope of application: sin, cos, log, exp Complexity for programmer: QoS_Compute, approx_loop extensions Semi-automatical Limited QoS approach Only numerical data as input, no structures in QoS modelling
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.