Personalized Diapause: Reducing Radio Energy Consumption of Smartphones by Network-Context Aware Dormancy Predictions

Size: px
Start display at page:

Download "Personalized Diapause: Reducing Radio Energy Consumption of Smartphones by Network-Context Aware Dormancy Predictions"

Transcription

1 1 A period of suspended growth accompanied by decreased metabolism in insects Personalized Diapause: Reducing Radio Energy Consumption of Smartphones by -Context Aware Dormancy Predictions Yeseong Kim and Jihong Kim Computer Architecture & Embedded Systems Lab. Department of Computer Science and Engineering Seoul National University 2012 Workshop on Power-Aware Computing and Systems October 7, 2012

2 2 Radio Energy Consumption in Smartphones High radio energy consumption About 30% of the total energy consumption in smartphones (3G ) Radio energy consumption increasing -dependent apps increasing (e.g., SNS apps) Shift to high-energy-demand 4G LTE radio network 3G or 4G LTE Apps Mobile Reducing radio energy consumption is an important design issue of smartphones.

3 Power (mw) 3 Radio Energy Consumption in Tail Time A significant portion of radio energy is consumed during tail time After a packet transmission, a high power level is maintained expecting a subsequent transmission. End of Transmission TAIL (10~20 Seconds) New Transmission No reconnection Time (ms)

4 Wasted tail energy of total radio energy (%) Power 4 Wasted Radio Energy during Tail Time A significant radio energy is wasted during tail time. If there is no transmission End of Transmission Wasted Tail Energy Tail Energy Tail time From our measurement study of 25 smartphone users 50% Time 40% 30% 20% 10% 0% 33% of the total radio energy is wasted in the tail time Reducing wasted radio energy is very important.

5 Power Packet Power Packet 5 Fast Dormancy and Key Challenge The Fast Dormancy feature enables a smartphone radio module to release the radio connection to save the wasted energy during the tail time Invoke fast dormancy Time Key Challenge: How to predict the subsequent transmission? Time

6 Power Packet Power Packet 6 Fast Dormancy and Key Challenge The Fast Dormancy feature enables a smartphone radio module to release the radio connection to save the wasted energy during the tail time Invoke fast dormancy Time Key Challenge: How to predict the subsequent transmission Right prediction: Saving radio energy Invoke fast dormancy Time

7 Power Packet Power Packet 7 Fast Dormancy and Key Challenge The Fast Dormancy feature enables a smartphone radio module to release the radio connection to save the wasted energy during the tail time Invoke fast dormancy Time Key Challenge: How to predict the subsequent transmission Wrong prediction: Additional Reconnection ; Long delay (e.g.,, 2 secs) to smartphone, Signaling overhead to mobile network Time

8 8 Problems of Existing Dormancy Technique Problem 1. Existing dormancy techniques are app-centric. require app-assisted run-time hints on the next transmission e.g., TOP [ICNP 10] Playing No more network access for a while App Developer Streaming No automatic App support Run-Time Hints User System Software

9 9 Problems of Existing Dormancy Technique Problem 2. Existing dormancy techniques are not applicable to most interactive apps. It is very difficult to predict how a user interacts with interactive apps such as google talk and facebook app. Mobile Interactive Apps Not applicable? User App Developer Run-Time Hints System Software

10 10 Contributions We propose Personalized Diapause, a general-purpose automatic predictive dormancy technique for supporting the fast dormancy feature Not depending on app-assisted future network usage hints Applicable to most of apps with general network transmission patterns Personalized Diapause was implemented on Android 2.3 (Gingerbread) Nexus S smartphones Radio energy consumption saving by up to 36% with 10% increase in the radio reconnection over when no fast dormancy feature is used.

11 11 Outline Introduction Overview of Personalized Diapause Key Steps of Personalized Diapause Extraction of Context Estimation of Transmission Trend Predictive Dormancy Analysis Experimental Results Conclusion

12 12 Key Idea of Personalized Diapause Activity Apps Sending a message Downloading a song Checking new s... Transmissions Transmissions Transmissions Transmissions Time

13 13 Key Idea of Personalized Diapause Activity Apps Sending a message Downloading a song Checking new s... Transmissions Context 1 Transmissions Context 2 Transmissions Context 3 Transmissions Context 4 Context Block Context Block Context Block Context 1 Context 3 Context 2 Context 4 Time

14 14 Key Idea of Personalized Diapause Personalized Usage Characteristics Activity Apps Transmission Trend Sending a message Downloading a song Checking new s... Transmission Trend Transmissions Context 1 Transmissions Context 2 Transmissions Context 3 Transmissions Context 4 Context Block Context Block Context Block Context 1 Context 3 Context 2 Context 4 Time

15 15 Overview of Personalized Diapause Key steps of Personalized Diapause Extracting semantically equivalent network activities + Estimating transmission trends of network activities Predictive dormancy analysis Context Block Context 1 Context 3 Transmission Trend Predictive Dormancy

16 16 Outline Introduction Overview of Personalized Diapause Key Steps of Personalized Diapause Extraction of Context Estimation of Transmission Trend Predictive Dormancy Analysis Experimental Results Conclusion

17 ~ Transmission 17 Step 1. Extraction of Context Transmissions are transferred due to network activities. Context Context Context Context Time Apps Downloading a song Downloading a song Fetching new s Sending an Activity

18 ~ Transmission 18 Step 1. Extraction of Context Transmissions are transferred due to network activities. Context Context Context Context Time Apps Downloading a song Downloading a song Fetching new s Sending an System Software Execution Path A Execution Path A Execution Path B Execution Path C The network activities can be systematically distinguished from their execution paths.

19 19 Identifying Equivalent Context Available from Dalvik VM connect, write, read, send, recv Socket API recv() recv() Transmission Time Downloading a song Context 1 Downloading a song Context 2

20 20 Identifying Equivalent Context Same execution paths Application Function Call Stack selectsong() makebuffer() downloadsong() selectsong() makebuffer() downloadsong() Socket API recv() recv() Transmission Time Context 1 Context 2 Context Block (NCB)

21 21 Context Block (NCB) contexts in the same Context Block are assumed to perform same network activity. Context 1 Context 2 Context 4 Downloading a song NCB 1 Context 3 Context 5 Context 6 Sending a message NCB 2 NCB 3 NCB 4 NCB 5 User s network context blocks Basic unit of monitoring transmission trend in the tail time

22 22 Key steps of Personalized Diapause Step 1. Step 2. Extracting semantically equivalent network activities + Estimating transmission trends of network activities Step 3. Predictive dormancy analysis Context Block Context 1 Context 3 Transmission Trend Predictive Dormancy

23 23 Personalized Transmission Trend Claim 1. Different users differently behave even for same network activity. Talk via messenger apps Mr. Every10Seconds Prof. EveryHour A transmission is likely to occur in the tail time A transmission is unlikely to occur in the tail time Must consider personalized transmission trends.

24 24 Transmission Trend of Activities Claim 2. Different Activities have different transmission trends. Checking system update Browsing a web page Sending a message Checking new s Transmission Trend Transmission Trend Transmission Trend Transmission Trend Must consider different network activity characteristics.

25 Transmission 25 Validation Study of Smartphone Usage Subject: 25 active smartphone users Aged 20~40 College students, graduate students, bankers, kindergarten teachers Study Period: during two weeks Method: using a modified Dalvik VM For logging network contexts with call stack information Context Whether/When did a next transmission occur in the tail time?? Tail time Time

26 User 1 User 2 User 3 User 4 User 5 User 6 User 7 User 8 User 9 User 10 User 11 User 12 User 13 User 14 User 15 User 16 User 17 User 18 User 19 User 20 User 21 User 22 User 23 User 24 User 25 Rate of occurrence of a next transmission in the tail time (%) 26 Personalized Usage Tendency Strong personalized network usage tendency Mr. Every10Seconds Prof. EveryHour Large energy wasted 0

27 Rate of occurrence of a next transmission (%) 27 Skewed Transmission Distribution User s behavior for transmissions is quite skewed Week 1 Week Most of the first transmissions in the tail happen within the first 6 seconds Time of occurrence of a next transmission (sec) These persistent right-skewed distribution can be exploited to apply the fast dormancy feature.

28 Rate of occurrence of a next transmission in the tail time (%) Transmission Characteristics for Different Activities Per User Different transmission characteristics for different network activities The transmission trend of each NCB persists over long time Checking system update Browsing a web page Sending a message Week 1 Week 2 Fetching new s Exploiting these persistent transmission trends over different NCBs,we can estimate transmissions in the tail time. 28

29 Step 2. Estimating Transmission Trend of Activity We estimate when/whether a transmission will occur in the tail time based on the skewed transmission distribution of each NCB. Context 1 A user s network context blocks Context 2 Context 4 Transmission Trend 29 Context 3 NCB 1 Context 5 Context 6 Transmission Trend NCB 2 Transmission Trend NCB 3 NCB 4 NCB 5 Transmission Trend Transmission Trend

30 30 Key steps of Personalized Diapause Step 1. Step 2. Extracting semantically equivalent network activities + Estimating transmission trends of network activities Step 3. Predictive dormancy analysis Context Block Context 1 Context 3 Transmission Trend Predictive Dormancy

31 Power 31 Step 3. Predictive Dormancy Analysis To determine when to invoke the fast dormancy feature Considering cost-benefit tradeoff Invoke fast dormancy at t i If a transmission occurs at t j (p j ) Tail Energy Canceled Benefit Over head Gain (G i ) = Expected Energy Benefit (B i ) Energy Cost (C j ) Time Probability (p j ) - ( ) Intuitively, choosing the best moment (t i ) to invoke fast dormancy Consider t k 's only where the probability of retransmissions after t k is less than a given upper bound threshold (See our paper for a detail description of the decision procedure) j

32 32 Architectural Overview of Personalized Diapause The key steps are implemented as additional modules to the Dalvik VM and Android framework. Call Stack Tracer Tail Time Power Model Application Dalvik VM Android Framework Personalized Activity Predictor Dormancy Granter Personalized Activity Predictor Context Block Extractor Immediate-Successor Trainer Cost-Benefit Analysis Engine

33 33 Outline Introduction Overview of Personalized Diapause Key Steps of Personalized Diapause Extraction of Context Estimation of Transmission Trend Predictive Dormancy Analysis Experimental Results Conclusion

34 34 Experimental Environment Implemented the Personalized Diapause (PD) technique on Nexus S Android reference smartphones Running Android 2.3 (Gingerbread) To Dalvik VM and Android framework Using the collected network transmission logs from 25 users A custom log replayer tool reproduced network contexts logs. A 3G energy simulator was used for energy consumption comparison. Log Replayer User Log Transmission & Fast dormancy Log 3G Energy Simulator Nexus S (Target Device)

35 Energy saving (%) 35 Impact of PD on Energy Consumption Saving Energy saving of Personalized Diapause No-fast-dormancy support as a baseline Reconnection increase limit 10% 15% 20% On average 60 23% energy saving with 10% of reconnection increase User 1 User 2 User 3 User 4 Mean (25 users)

36 Energy Saving (%) 36 Impact of NCB Classification Technique Comparison with Per-user PD Assuming that all network contexts are classified to a single NCB Per-NCB Per-user vs Per-user PD Per-user PD Very poor energy saving User 1 User 2 User 3 User 4 Mean (25 users) The fine-grained NCB separation based on semantic differences is important in achieving a high energy efficiency.

37 37 Conclusions We presented a general-purpose automatic predictive dormancy technique, Personalized Diapause. Optimizing the radio energy consumption of smartphones with the fast dormancy feature Personalized Diapause takes advantages of personalized network context usage in deciding when to release a radio connection. Based on an automatic extraction technique of meaningful network activities Future work Extend for other types of system optimizations using other useful information available from the network context.

38 Thank you 38

Personalized Diapause: Reducing Radio Energy Consumption of Smartphones by Network-Context Aware Dormancy Predictions

Personalized Diapause: Reducing Radio Energy Consumption of Smartphones by Network-Context Aware Dormancy Predictions Personalized Diapause: Reducing Radio Energy Consumption of Smartphones by Network-Context Aware Dormancy Predictions Yeseong Kim Jihong Kim Department of Computer Science and Engineering Seoul National

More information

Improving File System Performance of Mobile Storage Systems Using a Decoupled Defragmenter

Improving File System Performance of Mobile Storage Systems Using a Decoupled Defragmenter Improving File System Performance of Mobile Storage Systems Using a Decoupled Defragmenter Sangwook Shane Hahn *, Sungjin Lee, Cheng Ji, Li-Pin Chang +, Inhyuk Yee *, Liang Shi #, Chun Jason Xue and Jihong

More information

Understanding Storage I/O Behaviors of Mobile Applications. Louisiana State University Department of Computer Science and Engineering

Understanding Storage I/O Behaviors of Mobile Applications. Louisiana State University Department of Computer Science and Engineering Understanding Storage I/O Behaviors of Mobile Applications Jace Courville jcourv@csc.lsu.edu Feng Chen fchen@csc.lsu.edu Louisiana State University Department of Computer Science and Engineering The Rise

More information

Fine Grained Power Modeling For Smartphones Using System Call Tracing. Y. Charlie Hu Paramvir Bahl Yi-Min Wang

Fine Grained Power Modeling For Smartphones Using System Call Tracing. Y. Charlie Hu Paramvir Bahl Yi-Min Wang Fine Grained Power Modeling For Smartphones Using System Call Tracing Abhinav Pathak Ming Zhang Y. Charlie Hu Paramvir Bahl Yi-Min Wang 1 Smartphone is Energy Constrained Energy: One of the most critical

More information

Sense-Aid: A framework for enabling network as a service for participatory sensing

Sense-Aid: A framework for enabling network as a service for participatory sensing Sense-Aid: A framework for enabling network as a service for participatory sensing Heng Zhang Purdue ECE, Saurabh Bagchi Purdue ECE, He Wang Purdue CS, Rajesh K. Panta AT&T Labs CS logo Supported by: Slide

More information

L.C.Smith. Privacy-Preserving Offloading of Mobile App to the Public Cloud

L.C.Smith. Privacy-Preserving Offloading of Mobile App to the Public Cloud Privacy-Preserving Offloading of Mobile App to the Public Cloud Yue Duan, Mu Zhang, Heng Yin and Yuzhe Tang Department of EECS Syracuse University L.C.Smith College of Engineering 1 and Computer Science

More information

Migratory TCP (MTCP) Transport Layer Support for Highly Available Network Services

Migratory TCP (MTCP) Transport Layer Support for Highly Available Network Services Migratory TCP (MTCP) Transport Layer Support for Highly Available Network Services DisCo Lab Division of Computer and Information Sciences Rutgers University Nov. 29, 2001 CONS Light seminar 1 The People

More information

WiZi-Cloud: Application-transparent Dual ZigBee-WiFi Radios for Low Power Internet Access

WiZi-Cloud: Application-transparent Dual ZigBee-WiFi Radios for Low Power Internet Access WiZi-Cloud: Application-transparent Dual ZigBee-WiFi Radios for Low Power Internet Access Tao Jin, Guevara Noubir, Bo Sheng College of Computer and Information Science Northeastern University InfoCom 2011,

More information

RadioJockey: Mining Program Execution to Optimize Cellular Radio Usage

RadioJockey: Mining Program Execution to Optimize Cellular Radio Usage RadioJockey: Mining Program Execution to Optimize Cellular Radio Usage Pavan Kumar, Ranjita Bhagwan, Saikat Guha, Vishnu Navda, Ramachandran Ramjee, Dushyant Arora, Venkat Padmanabhan, George Varghese

More information

H.-S. Oh, B.-J. Kim, H.-K. Choi, S.-M. Moon. School of Electrical Engineering and Computer Science Seoul National University, Korea

H.-S. Oh, B.-J. Kim, H.-K. Choi, S.-M. Moon. School of Electrical Engineering and Computer Science Seoul National University, Korea H.-S. Oh, B.-J. Kim, H.-K. Choi, S.-M. Moon School of Electrical Engineering and Computer Science Seoul National University, Korea Android apps are programmed using Java Android uses DVM instead of JVM

More information

T Computer Networks Green ICT

T Computer Networks Green ICT T-110.4100 Computer Networks Green ICT 08.05.2012 Matti Siekkinen External sources: Y. Xiao: Green communications. T-110.5116 lecture. Aalto. 2010. Which one is Green ICT? Source: Google image What is

More information

PowerForecaster: Predicting Smartphone Power Impact of Continuous Sensing Applications at Pre-installation Time

PowerForecaster: Predicting Smartphone Power Impact of Continuous Sensing Applications at Pre-installation Time PowerForecaster: Predicting Smartphone Power Impact of Continuous Sensing Applications at Pre-installation Time Chulhong Min 1 Youngki Lee 2 Chungkuk Yoo 1 Seungwoo Kang 3 Sangwon Choi 1 Pillsoon Park

More information

Open Mobile Platforms. EE 392I, Lecture-6 May 4 th, 2010

Open Mobile Platforms. EE 392I, Lecture-6 May 4 th, 2010 Open Mobile Platforms EE 392I, Lecture-6 May 4 th, 2010 Open Mobile Platforms The Android Initiative T-Mobile s ongoing focus on Android based devices in US and EU markets In Nov 2007, Google announced

More information

Lecture 12: Model Serving. CSE599W: Spring 2018

Lecture 12: Model Serving. CSE599W: Spring 2018 Lecture 12: Model Serving CSE599W: Spring 2018 Deep Learning Applications That drink will get you to 2800 calories for today I last saw your keys in the store room Remind Tom of the party You re on page

More information

The Impact of Delay Variations on TCP Performance

The Impact of Delay Variations on TCP Performance INSTITUT FÜR KOMMUNIKATIONSNETZE UND RECHNERSYSTEME Prof. Dr.-Ing. Dr. h. c. mult. P. J. Kühn The Impact of Delay Variations on TCP Performance Michael Scharf scharf@ikr.uni-stuttgart.de ITG FG 5.2.1 Workshop,

More information

Predicting Messaging Response Time in a Long Distance Relationship

Predicting Messaging Response Time in a Long Distance Relationship Predicting Messaging Response Time in a Long Distance Relationship Meng-Chen Shieh m3shieh@ucsd.edu I. Introduction The key to any successful relationship is communication, especially during times when

More information

TANGO: Enabling New Services through Cooperation between Cellular Network and Mobile Devices. Motivation

TANGO: Enabling New Services through Cooperation between Cellular Network and Mobile Devices. Motivation TANGO: Enabling New Services through Cooperation between Cellular and Mobile Devices Nawanol Theera-Ampornpunt, Sambit Mishra, Saurabh Bagchi, Kaustubh Joshi (AT&T), Rajesh Panta (AT&T) Motivation Cellular

More information

Ubiquitous and Mobile Computing CS 528:EnergyEfficiency Comparison of Mobile Platforms and Applications: A Quantitative Approach. Norberto Luna Cano

Ubiquitous and Mobile Computing CS 528:EnergyEfficiency Comparison of Mobile Platforms and Applications: A Quantitative Approach. Norberto Luna Cano Ubiquitous and Mobile Computing CS 528:EnergyEfficiency Comparison of Mobile Platforms and Applications: A Quantitative Approach Norberto Luna Cano Computer Science Dept. Worcester Polytechnic Institute

More information

GreenBag: Energy-efficient Bandwidth Aggregation For Real-time Streaming in Heterogeneous Mobile Wireless Networks

GreenBag: Energy-efficient Bandwidth Aggregation For Real-time Streaming in Heterogeneous Mobile Wireless Networks GreenBag: Energy-efficient Bandwidth Aggregation For Real-time Streaming in Heterogeneous Mobile Wireless Networks Duc Hoang Bui, Kilho Lee, Sangeun Oh, Insik Shin Dept. of Computer Science KAIST, South

More information

Wireless Sensornetworks Concepts, Protocols and Applications. Chapter 5b. Link Layer Control

Wireless Sensornetworks Concepts, Protocols and Applications. Chapter 5b. Link Layer Control Wireless Sensornetworks Concepts, Protocols and Applications 5b Link Layer Control 1 Goals of this cha Understand the issues involved in turning the radio communication between two neighboring nodes into

More information

Computer Networks. Sándor Laki ELTE-Ericsson Communication Networks Laboratory

Computer Networks. Sándor Laki ELTE-Ericsson Communication Networks Laboratory Computer Networks Sándor Laki ELTE-Ericsson Communication Networks Laboratory ELTE FI Department Of Information Systems lakis@elte.hu http://lakis.web.elte.hu Based on the slides of Laurent Vanbever. Further

More information

Android framework. How to use it and extend it

Android framework. How to use it and extend it Android framework How to use it and extend it Android has got in the past three years an explosive growth: it has reached in Q1 2011 the goal of 100M of Activations world wide with a number of daily activations

More information

NESL. CAreDroid: Adaptation Framework for Android Context-Aware Applications. Salma Elmalaki Lucas Wanner Mani Srivastava

NESL. CAreDroid: Adaptation Framework for Android Context-Aware Applications. Salma Elmalaki Lucas Wanner Mani Srivastava CAreDroid: Adaptation Framework for Android Context-Aware Applications Salma Elmalaki Lucas Wanner Mani Srivastava 1 Isolated Disconnected Unaware Photo Courtesy: Student Portal 2 Computing From Isolation

More information

Survey Topic: WiFi On The Move Presented by - Abhinav Tekumalla (atekumal) Bahula Gupta (bahulag)

Survey Topic: WiFi On The Move Presented by - Abhinav Tekumalla (atekumal) Bahula Gupta (bahulag) Outline Survey Topic: WiFi On The Move Presented by - Abhinav Tekumalla (atekumal) Bahula Gupta (bahulag) WiFi on the move : Challenges VanLAN ViFi Cabernet HSPDA/ WiMax WiFi on the move Providing WiFi

More information

OUTLINE. Sharing videos. What is National Film Board (NFB)? Creating a free account Commenting on videos Creating a playlist

OUTLINE. Sharing videos. What is National Film Board (NFB)? Creating a free account Commenting on videos Creating a playlist NATIONAL FILM BOARD OUTLINE What is National Film Board (NFB)? Creating a free account Commenting on videos Creating a playlist Sharing videos Searching for videos Browsing for videos Finding similar videos

More information

Android App Development. Muhammad Sharjeel COMSATS Institute of Information Technology, Lahore

Android App Development. Muhammad Sharjeel COMSATS Institute of Information Technology, Lahore Android App Development Muhammad Sharjeel COMSATS Institute of Information Technology, Lahore Mobile devices (e.g., smartphone, tablet PCs, etc.) are increasingly becoming an essential part of human life

More information

Reliable Stream Analysis on the Internet of Things

Reliable Stream Analysis on the Internet of Things Reliable Stream Analysis on the Internet of Things ECE6102 Course Project Team IoT Submitted April 30, 2014 1 1. Introduction Team IoT is interested in developing a distributed system that supports live

More information

Introduction What is Android?

Introduction What is Android? Introduction What is Android? CS 2046 Mobile Application Development Fall 2010 Everything you know is wrong Most desktop/web applications: Large screen size vs. Everything you know is wrong Most desktop/web

More information

Traffic-aware techniques to reduce 3G/LTE wireless energy consumption

Traffic-aware techniques to reduce 3G/LTE wireless energy consumption Traffic-aware techniques to reduce 3G/LTE wireless energy consumption The MIT Faculty has made this article openly available. Please share how this access benefits you. Your story matters. Citation As

More information

Testing & Assuring Mobile End User Experience Before Production Neotys

Testing & Assuring Mobile End User Experience Before Production Neotys Testing & Assuring Mobile End User Experience Before Production Neotys Henrik Rexed Agenda Introduction The challenges Best practices NeoLoad mobile capabilities Mobile devices are used more and more At

More information

University Bulletin Board Application

University Bulletin Board Application University Bulletin Board Application Introduction In many universities and colleges there are many bulletin boards or notice boards filled with fliers that contain information on seminars, events, selling

More information

Measuring the impact of IoT. Alison Robart Director, Client Services

Measuring the impact of IoT. Alison Robart Director, Client Services Measuring the impact of IoT Alison Robart Director, Client Services Looking Ahead to the Voice Era July 2017 Alison Robart Director Client Insights comscore, Inc. For info about the proprietary technology

More information

Mandatory Access Control for the Android Dalvik VM

Mandatory Access Control for the Android Dalvik VM Mandatory Access Control for the Android Dalvik VM ESOS 13 Aline Bousquet, Jérémy Briffaut, Laurent Clevy, Christian Toinard, Benjamin Venelle June 25, 2013 Esos 13 Mandatory Access Control for the Android

More information

CloneCloud: Elastic Execution between Mobile Device and Cloud, Chun et al.

CloneCloud: Elastic Execution between Mobile Device and Cloud, Chun et al. CloneCloud: Elastic Execution between Mobile Device and Cloud, Chun et al. Noah Apthorpe Department of Computer Science Princeton University October 14th, 2015 Noah Apthorpe CloneCloud 1/16 Motivation

More information

A First Look at Traffic on Smartphones

A First Look at Traffic on Smartphones A First Look at Traffic on Smartphones by Falaki et al. Andrew Zafft CS Department Agenda Objective Study Structure Outcomes & Observations Future Work / Citations Conclusions 2 Objective Statistics Why

More information

CSC8223 Wireless Sensor Networks. Chapter 3 Network Architecture

CSC8223 Wireless Sensor Networks. Chapter 3 Network Architecture CSC8223 Wireless Sensor Networks Chapter 3 Network Architecture Goals of this chapter General principles and architectures: how to put the nodes together to form a meaningful network Design approaches:

More information

Using Freegal on an Android Device

Using Freegal on an Android Device Using Freegal on an Android Device What is Freegal? 2 Download the Freegal App 2 Set Up the Freegal App 3 Email Notifications 4 Find a Song, Artist or Album to Listen To 4 My Music 5 Stream Music 6 Download

More information

IX: A Protected Dataplane Operating System for High Throughput and Low Latency

IX: A Protected Dataplane Operating System for High Throughput and Low Latency IX: A Protected Dataplane Operating System for High Throughput and Low Latency Belay, A. et al. Proc. of the 11th USENIX Symp. on OSDI, pp. 49-65, 2014. Reviewed by Chun-Yu and Xinghao Li Summary In this

More information

Understanding the Characteristics of Android Wear OS. Renju Liu and Felix Xiaozhu Lin Purdue ECE

Understanding the Characteristics of Android Wear OS. Renju Liu and Felix Xiaozhu Lin Purdue ECE Understanding the Characteristics of Android Wear OS Renju Liu and Felix Xiaozhu Lin Purdue ECE The Wearable stack 5 Top questions Wearables should enjoy Baremetal performance Baremetal efficiency In this

More information

THE ONLINER A VIP FOR MARKETERS. Slovenia

THE ONLINER A VIP FOR MARKETERS. Slovenia I THE ONLINER A VIP FOR MARKETERS Slovenia I INTERNET USAGE Internet usage is still growing slightly and is now close to the saturation point. Underlying developments such as a higher daily reach are stemming

More information

Improving Localization and Energy Efficiency of Smartphone Applications

Improving Localization and Energy Efficiency of Smartphone Applications Improving Localization and Energy Efficiency of Smartphone Applications Swadhin Pradhan Supervisor: Niloy Ganguly Department of Computer Science & Engineering Indian Institute of Technology Kharagpur MS

More information

Method-Level Phase Behavior in Java Workloads

Method-Level Phase Behavior in Java Workloads Method-Level Phase Behavior in Java Workloads Andy Georges, Dries Buytaert, Lieven Eeckhout and Koen De Bosschere Ghent University Presented by Bruno Dufour dufour@cs.rutgers.edu Rutgers University DCS

More information

Energy-Aware CPU Frequency Scaling for Mobile Video Streaming

Energy-Aware CPU Frequency Scaling for Mobile Video Streaming 1 Energy-Aware CPU Frequency Scaling for Mobile Video Streaming Yi Yang, Student Member, IEEE, Wenjie Hu, Student Member, IEEE, Xianda Chen, Student Member, IEEE, Guohong Cao, Fellow, IEEE, Abstract The

More information

Ad hoc and Sensor Networks Chapter 3: Network architecture

Ad hoc and Sensor Networks Chapter 3: Network architecture Ad hoc and Sensor Networks Chapter 3: Network architecture Holger Karl Computer Networks Group Universität Paderborn Computer Networks Group Universität Paderborn Outline Design principles (skipped) Basic

More information

Ad hoc and Sensor Networks Chapter 3: Network architecture

Ad hoc and Sensor Networks Chapter 3: Network architecture Ad hoc and Sensor Networks Chapter 3: Network architecture Holger Karl Computer Networks Group Universität Paderborn Goals of this chapter Having looked at the individual nodes in the previous chapter,

More information

Ad hoc and Sensor Networks Chapter 3: Network architecture

Ad hoc and Sensor Networks Chapter 3: Network architecture Ad hoc and Sensor Networks Chapter 3: Network architecture Holger Karl, Andreas Willig, "Protocols and Architectures for Wireless Sensor Networks," Wiley 2005 Goals of this chapter Having looked at the

More information

Ad hoc and Sensor Networks Chapter 3: Network architecture

Ad hoc and Sensor Networks Chapter 3: Network architecture Ad hoc and Sensor Networks Chapter 3: Network architecture Goals of this chapter Having looked at the individual nodes in the previous chapter, we look at general principles and architectures how to put

More information

Interference-Aware Real-Time Flow Scheduling for Wireless Sensor Networks

Interference-Aware Real-Time Flow Scheduling for Wireless Sensor Networks Interference-Aware Real-Time Flow Scheduling for Wireless Sensor Networks Octav Chipara, Chengjie Wu, Chenyang Lu, William Griswold University of California, San Diego Washington University in St. Louis

More information

University of Maryland at College Park Department of Geographical Sciences GEOG 477/ GEOG777: Mobile GIS Development

University of Maryland at College Park Department of Geographical Sciences GEOG 477/ GEOG777: Mobile GIS Development University of Maryland at College Park Department of Geographical Sciences GEOG 477/ GEOG777: Mobile GIS Development Instructor: Dr. Ruibo Han Office: LeFrak Hall (LEF) 1111B Email: ruibo@umd.edu (preferred)

More information

Tuning RED for Web Traffic

Tuning RED for Web Traffic Tuning RED for Web Traffic Mikkel Christiansen, Kevin Jeffay, David Ott, Donelson Smith UNC, Chapel Hill SIGCOMM 2000, Stockholm subsequently IEEE/ACM Transactions on Networking Vol. 9, No. 3 (June 2001)

More information

The State of the User Experience

The State of the User Experience The State of the User Experience 2014 Annual Edition Courtesy of Table of Contents Executive Summary............................2 Key Findings................................4 Figures Figure 1: Rank in

More information

Project Details. Jiasi Chen CS 179i: Project in Computer Science (Networks) Lectures: Monday 3:10-4pm in Spieth 1307

Project Details. Jiasi Chen CS 179i: Project in Computer Science (Networks) Lectures: Monday 3:10-4pm in Spieth 1307 Project Details Jiasi Chen CS 179i: Project in Computer Science (Networks) Lectures: Monday 3:10-4pm in Spieth 1307 http://www.cs.ucr.edu/~jiasi/teaching/cs179i_winter16/ 1 Outline Virtual reality Video

More information

Advanced Computer Networks. Wireless TCP

Advanced Computer Networks. Wireless TCP Advanced Computer Networks 263 3501 00 Wireless TCP Patrick Stuedi Spring Semester 2014 1 Oriana Riva, Department of Computer Science ETH Zürich Outline Last week: Today: Cellular Networks Mobile IP Wireless

More information

An AI-Assisted Cyber Attack Detection Framework for Software Defined Mobile Networks

An AI-Assisted Cyber Attack Detection Framework for Software Defined Mobile Networks An AI-Assisted Cyber Attack Detection Framework for Software Defined Mobile Networks G. Catania 1, L. Ganga 1, S. Milardo 2, G. Morabito 3, A. Mursia 1 1 Land & Naval Defence Electronics Division - Leonardo

More information

Introduction to Internet of Things Prof. Sudip Misra Department of Computer Science and Engineering Indian Institute of Technology, Kharagpur

Introduction to Internet of Things Prof. Sudip Misra Department of Computer Science and Engineering Indian Institute of Technology, Kharagpur Introduction to Internet of Things Prof. Sudip Misra Department of Computer Science and Engineering Indian Institute of Technology, Kharagpur Lecture - 35 Software-Defined lot Networking - Part- 1 Having

More information

Low Latency via Redundancy

Low Latency via Redundancy Low Latency via Redundancy Ashish Vulimiri, Philip Brighten Godfrey, Radhika Mittal, Justine Sherry, Sylvia Ratnasamy, Scott Shenker Presenter: Meng Wang 2 Low Latency Is Important Injecting just 400 milliseconds

More information

Today s Topics. The Problem. MAUI: Making Smartphones Last Longer With Code Offload 10/10/2015. Battery fails to keep up. The Problem.

Today s Topics. The Problem. MAUI: Making Smartphones Last Longer With Code Offload 10/10/2015. Battery fails to keep up. The Problem. The Problem Motivation Evaluation : Making Smartphones Last Longer With Code Offload Slides based on a paper by: Eduardo Cuervo (Duke University), Aruna Balasubramanian (UMass. Amherst), Dae-ki Cho (UCLA),

More information

FlexiWeb: Network-Aware Compaction for Accelerating Mobile Web

FlexiWeb: Network-Aware Compaction for Accelerating Mobile Web FlexiWeb: Network-Aware Compaction for Accelerating Mobile Web What s the impact of web latency? 100ms 1% Delay sales Source : https://speakerdeck.com/deanohume/faster-mobilewebsites! 100ms 1% Delay revenue

More information

Building Consistent Transactions with Inconsistent Replication

Building Consistent Transactions with Inconsistent Replication DB Reading Group Fall 2015 slides by Dana Van Aken Building Consistent Transactions with Inconsistent Replication Irene Zhang, Naveen Kr. Sharma, Adriana Szekeres, Arvind Krishnamurthy, Dan R. K. Ports

More information

Building a Profitable Data Future. Monetizing Data Traffic

Building a Profitable Data Future. Monetizing Data Traffic Building a Profitable Data Future Monetizing Data Traffic Market Introduction Driven by an upsurge in smartphone usage, increased mobile-broadband penetration and enormous growth in video consumption,

More information

Schema-Agnostic Indexing with Azure Document DB

Schema-Agnostic Indexing with Azure Document DB Schema-Agnostic Indexing with Azure Document DB Introduction Azure DocumentDB is Microsoft s multi-tenant distributed database service for managing JSON documents at Internet scale Multi-tenancy is an

More information

Intelligent Edge Computing and ML-based Traffic Classifier. Kwihoon Kim, Minsuk Kim (ETRI) April 25.

Intelligent Edge Computing and ML-based Traffic Classifier. Kwihoon Kim, Minsuk Kim (ETRI)  April 25. Intelligent Edge Computing and ML-based Traffic Classifier Kwihoon Kim, Minsuk Kim (ETRI) (kwihooi@etri.re.kr, mskim16@etri.re.kr) April 25. 2018 ITU Workshop on Impact of AI on ICT Infrastructures Cian,

More information

Implementation Experiments on HighSpeed and Parallel TCP

Implementation Experiments on HighSpeed and Parallel TCP Implementation Experiments on HighSpeed and TCP Zongsheng Zhang Go Hasegawa Masayuki Murata Osaka University Outline Introduction Background of and g Why to evaluate in a test-bed network A refined algorithm

More information

The War Between Mice and Elephants

The War Between Mice and Elephants The War Between Mice and Elephants Liang Guo and Ibrahim Matta Computer Science Department Boston University 9th IEEE International Conference on Network Protocols (ICNP),, Riverside, CA, November 2001.

More information

Project 0: Implementing a Hash Table

Project 0: Implementing a Hash Table Project : Implementing a Hash Table CS, Big Data Systems, Spring Goal and Motivation. The goal of Project is to help you refresh basic skills at designing and implementing data structures and algorithms.

More information

Multimedia Streaming. Mike Zink

Multimedia Streaming. Mike Zink Multimedia Streaming Mike Zink Technical Challenges Servers (and proxy caches) storage continuous media streams, e.g.: 4000 movies * 90 minutes * 10 Mbps (DVD) = 27.0 TB 15 Mbps = 40.5 TB 36 Mbps (BluRay)=

More information

CROSS LAYER PROTOCOL (APTEEN) USING WSN FOR REAL TIME APPLICATION

CROSS LAYER PROTOCOL (APTEEN) USING WSN FOR REAL TIME APPLICATION CROSS LAYER PROTOCOL (APTEEN) USING WSN FOR REAL TIME APPLICATION V. A. Dahifale 1, N. Y. Siddiqui 2 PG Student, College of Engineering Kopargaon, Maharashtra, India 1 Assistant Professor, College of Engineering

More information

Android project proposals

Android project proposals Android project proposals Luca Bedogni (lbedogni@cs.unibo.it) April 7, 2016 Introduction In this document, we describe four possible projects for the exam of the Laboratorio di applicazioni mobili course.

More information

Granting Silence to Avoid Wireless Collisions

Granting Silence to Avoid Wireless Collisions Granting Silence to Avoid Wireless Collisions Jung Il Choi, Mayank Jain, Maria A. Kazandjieva, and Philip Levis October 6, 2010 ICNP 2010 Wireless Mesh and CSMA One UDP flow along a static 4-hop route

More information

The UK Online Audience. Julie Forey IAB Research Breakfast July 2018

The UK Online Audience. Julie Forey IAB Research Breakfast July 2018 The UK Online Audience Julie Forey IAB Research Breakfast July 2018 UKOM Insights: The Ozone Project The 3 News brands have a combined audience of 39.7m Total Unique Visitors 18+ (000) % Reach 18 + Digital

More information

Project 0: Implementing a Hash Table

Project 0: Implementing a Hash Table CS: DATA SYSTEMS Project : Implementing a Hash Table CS, Data Systems, Fall Goal and Motivation. The goal of Project is to help you develop (or refresh) basic skills at designing and implementing data

More information

High-performance and Low-power Consumption Vector Processor for LTE Baseband LSI

High-performance and Low-power Consumption Vector Processor for LTE Baseband LSI High-performance and Low-power Consumption Vector Processor for LTE Baseband LSI Yi Ge Mitsuru Tomono Makiko Ito Yoshio Hirose Recently, the transmission rate for handheld devices has been increasing by

More information

MTAT Enterprise System Integration. Lecture 2: Middleware & Web Services

MTAT Enterprise System Integration. Lecture 2: Middleware & Web Services MTAT.03.229 Enterprise System Integration Lecture 2: Middleware & Web Services Luciano García-Bañuelos Slides by Prof. M. Dumas Overall view 2 Enterprise Java 2 Entity classes (Data layer) 3 Enterprise

More information

Mobile Transport Layer

Mobile Transport Layer Mobile Transport Layer 1 Transport Layer HTTP (used by web services) typically uses TCP Reliable transport between TCP client and server required - Stream oriented, not transaction oriented - Network friendly:

More information

Mobile Communications Chapter 9: Mobile Transport Layer

Mobile Communications Chapter 9: Mobile Transport Layer Prof. Dr.-Ing Jochen H. Schiller Inst. of Computer Science Freie Universität Berlin Germany Mobile Communications Chapter 9: Mobile Transport Layer Motivation, TCP-mechanisms Classical approaches (Indirect

More information

Android project proposals

Android project proposals Android project proposals Luca Bedogni Marco Di Felice ({lbedogni,difelice}@cs.unibo.it) May 2, 2014 Introduction In this document, we describe four possible projects for the exam of the Laboratorio di

More information

Sub-millisecond Stateful Stream Querying over Fast-evolving Linked Data

Sub-millisecond Stateful Stream Querying over Fast-evolving Linked Data Sub-millisecond Stateful Stream Querying over Fast-evolving Linked Data Yunhao Zhang, Rong Chen, Haibo Chen Institute of Parallel and Distributed Systems (IPADS) Shanghai Jiao Tong University Stream Query

More information

Energy-Aware Web Browsing on Smartphones. Bo Zhao, Wenjie Hu, Qiang Zheng, and Guohong Cao

Energy-Aware Web Browsing on Smartphones. Bo Zhao, Wenjie Hu, Qiang Zheng, and Guohong Cao information: DOI 1.119/TPDS.214.2312931, IEEE Transactions on Parallel and Distributed Systems 1 Energy-Aware Web Browsing on Smartphones Bo Zhao, Wenjie Hu, Qiang Zheng, and Guohong Cao Abstract Smartphone

More information

Suspend-aware Segment Cleaning in Log-Structured File System

Suspend-aware Segment Cleaning in Log-Structured File System USENI HotStorage 15 Santa Clara, CA, USA, July 6~7, 2015 Suspend-aware Segment Cleaning in Log-Structured File System Dongil Park, Seungyong Cheon, Youjip Won Hanyang University Outline Introduction Log-structured

More information

Chapter 2 Application Layer. Lecture 4: principles of network applications. Computer Networking: A Top Down Approach

Chapter 2 Application Layer. Lecture 4: principles of network applications. Computer Networking: A Top Down Approach Chapter 2 Application Layer Lecture 4: principles of network applications Computer Networking: A Top Down Approach 6 th edition Jim Kurose, Keith Ross Addison-Wesley March 2012 Application Layer 2-1 Chapter

More information

Why Android? Why Android? Android Overview. Why Mobile App Development? 20-Nov-18

Why Android? Why Android? Android Overview. Why Mobile App Development? 20-Nov-18 Why Android? Android Overview Dr. Siddharth Kaza Dr. Josh Dehlinger A lot of students have them 2010 survey by University of CO 1 : 22% of college students have Android phone (26% Blackberry, 40% iphone)

More information

DefDroid: Towards a More Defensive Mobile OS Against Disruptive App Behavior

DefDroid: Towards a More Defensive Mobile OS Against Disruptive App Behavior http://defdroid.org DefDroid: Towards a More Defensive Mobile OS Against Disruptive App Behavior Peng (Ryan) Huang, Tianyin Xu, Xinxin Jin, Yuanyuan Zhou UC San Diego Growing number of (novice) app developers

More information

Proceedings of the Fourth Engineering Students Conference at Peradeniya (ESCaPe) SDN Flow Caching

Proceedings of the Fourth Engineering Students Conference at Peradeniya (ESCaPe) SDN Flow Caching Proceedings of the Fourth Engineering Students Conference at Peradeniya (ESCaPe) 2016 SDN Flow Caching N.B.U.S. Nanayakkara, R.M.L.S. Bandara, N.B. Weerasinghe, S,N, Karunarathna Department of Computer

More information

A Machine Learning Based Approach to Mobile Network Analysis

A Machine Learning Based Approach to Mobile Network Analysis A Machine Learning Based Approach to Mobile Network Analysis Zengwen Yuan 1, Yuanjie Li 1, Chunyi Peng 2, Songwu Lu 1, Haotian Deng 2, Zhaowei Tan 1, Taqi Raza 1 1 2 Overview 2 Overview 2 Background Overview

More information

Software defined radio networking: Opportunities and challenges

Software defined radio networking: Opportunities and challenges Software defined radio networking: Opportunities and challenges Navid Nikaein Putting more IT/SW to the network EURECOM, Mobile Communication Department Eurecom Graduate school and research center in the

More information

Network-Adaptive Video Coding and Transmission

Network-Adaptive Video Coding and Transmission Header for SPIE use Network-Adaptive Video Coding and Transmission Kay Sripanidkulchai and Tsuhan Chen Department of Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh, PA 15213

More information

Research Project. SUBASIC RESEARCH! Jake Pflum! Stephanie Martinez! Richard Escobar! Anh Nguyen! Ajla Subasic

Research Project. SUBASIC RESEARCH! Jake Pflum! Stephanie Martinez! Richard Escobar! Anh Nguyen! Ajla Subasic Research Project SUBASIC RESEARCH Jake Pflum Stephanie Martinez Richard Escobar Anh Nguyen Ajla Subasic Table of Contents Statement of the Problem 3 Literature Review 4 Research Questions & Hypotheses

More information

Bluetooth Smart: How to avoid making dumb devices

Bluetooth Smart: How to avoid making dumb devices Bluetooth Smart: How to avoid making dumb devices CONNECTED DEVICES Bluetooth Smart: How to avoid making dumb devices white paper Development activity for new Bluetooth Smart devices rocketed after Apple

More information

IP Mobility vs. Session Mobility

IP Mobility vs. Session Mobility IP Mobility vs. Session Mobility Securing wireless communication is a formidable task, something that many companies are rapidly learning the hard way. IP level solutions become extremely cumbersome when

More information

Slide 1. Opera Max. Migrating the next billion smartphone users for better app experience

Slide 1. Opera Max. Migrating the next billion smartphone users for better app experience Slide 1 Opera Max Migrating the next billion smartphone users for better app experience The 3 Consideration in the Next Billion Migration Slide 2 Cost of ownership (Device) Cost of usage (Data) Network

More information

Operating Systems. Operating Systems Sina Meraji U of T

Operating Systems. Operating Systems Sina Meraji U of T Operating Systems Operating Systems Sina Meraji U of T Recap Last time we looked at memory management techniques Fixed partitioning Dynamic partitioning Paging Example Address Translation Suppose addresses

More information

Supporting Service Differentiation for Real-Time and Best-Effort Traffic in Stateless Wireless Ad-Hoc Networks (SWAN)

Supporting Service Differentiation for Real-Time and Best-Effort Traffic in Stateless Wireless Ad-Hoc Networks (SWAN) Supporting Service Differentiation for Real-Time and Best-Effort Traffic in Stateless Wireless Ad-Hoc Networks (SWAN) G. S. Ahn, A. T. Campbell, A. Veres, and L. H. Sun IEEE Trans. On Mobile Computing

More information

The Nielsen Comparable Q2 2016

The Nielsen Comparable Q2 2016 The Nielsen Comparable Metrics Report Q2 2016 The Comparable Metrics Series Q2 2016 Copyright 2016 The Nielsen Company 1 welcome Welcome to the Q2 2016 Nielsen Comparable Metrics Report! This is an in-depth

More information

Multithreaded Processors. Department of Electrical Engineering Stanford University

Multithreaded Processors. Department of Electrical Engineering Stanford University Lecture 12: Multithreaded Processors Department of Electrical Engineering Stanford University http://eeclass.stanford.edu/ee382a Lecture 12-1 The Big Picture Previous lectures: Core design for single-thread

More information

Sukjun Lim Strategic Planning, User interaction, Design research specialist

Sukjun Lim Strategic Planning, User interaction, Design research specialist Sukjun Lim Strategic Planning, User interaction, Design research specialist LINE MUSIC Strategic Planning and Interaction Design Convergence LINE MUSIC is an on-line music streaming service. Japan has

More information

modern database systems lecture 10 : large-scale graph processing

modern database systems lecture 10 : large-scale graph processing modern database systems lecture 1 : large-scale graph processing Aristides Gionis spring 18 timeline today : homework is due march 6 : homework out april 5, 9-1 : final exam april : homework due graphs

More information

Opportunistic Spectrum Usage: Bounds and a Multi-band CSMA/CA Protocol. Ashu Sabharwal, Vikram Kanodia and Ed Knightly Rice University

Opportunistic Spectrum Usage: Bounds and a Multi-band CSMA/CA Protocol. Ashu Sabharwal, Vikram Kanodia and Ed Knightly Rice University Opportunistic Spectrum Usage: Bounds and a Multi-band CSMA/CA Protocol, Vikram Kanodia and Ed Knightly ECE, High throughput High availability Economic viability Commonly Held Vision Killer app is the network

More information

Cellular Networks and Mobile Compu5ng COMS , Spring 2012

Cellular Networks and Mobile Compu5ng COMS , Spring 2012 Cellular Networks and Mobile Compu5ng COMS 6998-8, Spring 2012 Instructor: Li Erran Li (lierranli@cs.columbia.edu) hkp://www.cs.columbia.edu/~coms6998-8/ 2/27/2012: Radio Resource Usage Profiling and Op5miza5on

More information

Wireless Internet Routing. Learning from Deployments Link Metrics

Wireless Internet Routing. Learning from Deployments Link Metrics Wireless Internet Routing Learning from Deployments Link Metrics 1 Learning From Deployments Early worked focused traditional routing issues o Control plane: topology management, neighbor discovery o Data

More information

Internet Quality of Service: an Overview

Internet Quality of Service: an Overview Internet Quality of Service: an Overview W. Zhao and et al, Columbia University presented by 리준걸 2006.10.25 INC Lab, Seoul Nat l University Outline Introduce QoS framework IntServ DiffServ Detailed mechanism

More information