phd pre-thesis

15
Analyzing and Optimizing Abnormal Energy Consumption in Software Systems Rui Pereira [email protected] Universidade do Minho João Saraiva [email protected] Universidade do Minho Jácome Cunha [email protected] Universidade Nova de Lisboa Supervisors Pre-Thesis

Upload: rui-pereira

Post on 28-Jun-2015

98 views

Category:

Presentations & Public Speaking


2 download

DESCRIPTION

Analyzing and Optimizing Abnormal Energy Consumption in Software Systems

TRANSCRIPT

Page 1: PhD Pre-Thesis

Analyzing and Optimizing Abnormal Energy Consumption in Software Systems

Rui Pereira

[email protected]

Universidade do Minho

João Saraiva

[email protected]

Universidade do Minho

Jácome Cunha

[email protected]

Universidade Nova de Lisboa

Supervisors

Pre-Thesis

Page 2: PhD Pre-Thesis

Going Green

+ =

1Pre-Thesis

Page 3: PhD Pre-Thesis

•Caught the attention of many companies allowing them to save:

Green Computing

2

“close to 50% of the energy costs of an organization can be attributed to the IT departments”- [Harmon and Auseklis, 2009]

Pre-Thesis

Page 4: PhD Pre-Thesis

Green Computing – Greenness by IT

3

“up to 90% of energy used by ICT hardware can be attributed to software”

- [Standard, 2013]

Pre-Thesis

Page 5: PhD Pre-Thesis

Green Computing – Greenness of IT

4Pre-Thesis

Page 6: PhD Pre-Thesis

PhD Research Motivation

5

PhD

Mining questions about software energy consumption- [Pinto et al., 2014]

Mining energy-greedy API usage patterns in android apps: An empirical study- [Linares-Vasquez et al., 2014]

Seeds: A software engineer’s energy-optimization decision support framework- [Manotas et al., 2014]

Calculating source line level energy information for android applications- [Li et al., 2013]

Initial explorations on design pattern energy usage- [Sahin et al., 2012]

How does code obfuscation impact energy usage?- [Sahin et al., 2014]

Pre-Thesis

Page 7: PhD Pre-Thesis

PhD Research Questions

6

1. Is it possible to accurately detect and identify energy faults? 2. Can these energy faults be optimized without the expense of

program functionality?3. Can we show this information in a manner allowing programmers to

become energy-aware while programming?

Pre-Thesis

Page 8: PhD Pre-Thesis

PhD Research Plan

7Pre-Thesis

Page 9: PhD Pre-Thesis

PhD Research Plan - Analysis

8

TestsCo

mpo

nent

s

1

t1 t2 t3 t4 t5

1

1111

1

1100

1 1

1 1 1

1 1 11 11 0 11 0 1

1 1 1

0.30.7

0.3

0.3

0.3

0.7

1 0 1 0 1

Pre-Thesis

Page 10: PhD Pre-Thesis

PhD Research Plan - Analysis

9Pre-Thesis

Page 11: PhD Pre-Thesis

PhD Research Plan - Visualization

10Pre-Thesis

Page 12: PhD Pre-Thesis

PhD Research Plan - Optimization

11Pre-Thesis

Page 13: PhD Pre-Thesis

Industrial Partners

12Pre-Thesis

Page 14: PhD Pre-Thesis

Past Year

13

•Grand Prize Winners ($2000) Energy Consumption Detection in LabVIEW -

•Visit to - +

•Publications:• Graphical querying of model-driven spreadsheets (HCI’14)

• Refactoring smelly spreadsheet models (ICCSA’14)

• Embedding model-driven spreadsheet queries in spreadsheet systems (VL/HCC’14)

• ES-SQL: Visually querying spreadsheets (VL/HCC’14)

• Refactoring meets model-driven spreadsheet evolution (QUATIC’14)

Pre-Thesis

Page 15: PhD Pre-Thesis

Analyzing and Optimizing Abnormal Energy Consumption in Software Systems

Rui Pereira

[email protected]

Universidade do Minho

João Saraiva

[email protected]

Universidade do Minho

Jácome Cunha

[email protected]

Universidade Nova de Lisboa

Supervisors

Pre-Thesis