Client-Side Energy Efficiency of HTTP/2 for Web and Mobile App Developers

Chowdhury, Shaiful Alam; Sapra, Varun; Hindle, Abram

doi:10.1109/saner.2016.77

Cited by 24 publications

(21 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To understand the effect size, we used Cohen's d. Cohen's d compares the means of the two normallydistributed samples, taking in to account their pooled standard deviation to determine the offset [5]. Larger [4] that simply using SSL/TLS has a significant effect on energy consumption. The difference between bare-metal Linux versus enabling SSL on the same configuration also has a large effect size, with Cohen's d calculated to be 1.32.…”

Section: Postgresqlmentioning

confidence: 99%

How does docker affect energy consumption? Evaluating workloads in and out of Docker containers

Santos

McLean

Solinas

et al. 2018

Journal of Systems and Software

Self Cite

View full text Add to dashboard Cite

Context: Virtual machines provide isolation of services at the cost of hypervisors and more resource usage. This spurred the growth of systems like Docker that enable single hosts to isolate several applications, similar to VMs, within a low-overhead abstraction called containers. Motivation: Although containers tout low overhead performance, do they still have low energy consumption? Methodology: This work statistically compares (ttest, Wilcoxon) the energy consumption of three application workloads in Docker and on bare-metal Linux. Results: In all cases, there was a statistically significant (t-test and Wilcoxon p < 0.05) increase in energy consumption when running tests in Docker, mostly due to the performance of I/O system calls.

show abstract

Section: Postgresqlmentioning

confidence: 99%

How does docker affect energy consumption? Evaluating workloads in and out of Docker containers

Santos

McLean

Solinas

et al. 2018

Journal of Systems and Software

Self Cite

View full text Add to dashboard Cite

show abstract

“…Power (P) is defined as the rate of work completion and measured in watts whereas energy (E) is the total amount of work done for a given time and expressed in joules [5,10,11]. The difference between power and energy, E = P · T , is one of the first requirement to understand in order to develop energy efficient system.…”

Section: Power Is Not Energymentioning

confidence: 99%

“…An energy bug-in contrast to the traditional programming bugs producing incorrect/unwanted resultsspecifically leads to reduced battery life of battery-driven devices [56]. The sources of energy bugs can be of different types: no sleep bug-result of waking up the CPU, but not putting back to sleep mode [35]; loop bug-waiting for an event to happen, and thus periodically inspecting changes in a variable [43]; and tail energy leak-observed with some hardware components that do not stop drawing power immediately after job completion [11,44].…”

Section: Energy Bugsmentioning

confidence: 99%

“…Another source of energy consumption is tail energy leaks. Bundling I/O operations together was found to be very effective to reduce the number of tail energy leaks [45,11,32]. Interestingly enough, energy consumption of smart-phones can be effected by the type of servers they communicate with.…”

Section: Energy Optimization and Guidelinesmentioning

confidence: 99%

“…Interestingly enough, energy consumption of smart-phones can be effected by the type of servers they communicate with. Chowdhury et al observed that employing HTTP/2 server can help significantly in reducing clients energy consumption [11]. As screen color is very sensitive for OLED screen's energy consumption, tools for automatic color transformation have been developed [33,34].…”

Section: Energy Optimization and Guidelinesmentioning

confidence: 99%

See 2 more Smart Citations

Did I make a mistake? Finding the impact of code change on energy regression

Chowdhury

Gil

Romansky

et al. 2017

Preprint

Self Cite

View full text Add to dashboard Cite

Software energy consumption is a performance related non-functional requirement that complicates building software on mobile devices today. Energy hogging applications are a liability to both the end-user and software developer. Measuring software energy consumption is non-trivial, requiring both equipment and expertise, yet many researchers have found that software energy consumption can be modelled. Prior works have hinted that with more energy measurement data one can make more accurate energy models but this data was expensive to extract because it required energy measurement of running test cases (rare) or time consuming manually written tests. We address these concerns by automatically generating test cases to drive applications undergoing energy measurement. Automatic test generation allows a model to be continuously improved in a model building process whereby applications are extracted, tests are generated, energy is measured and combined with instrumentation to train a grander big-data model of software energy consumption. This continuous process has allowed the authors to generate and extract measurements from hundreds of applications in order to build accurate energy models capable of predicting the energy consumption of applications without end-user energy measurement. We clearly show that models built from more applications reduce energy modelling error.

show abstract