Transcript lec13-summary

1
Cellular Networks and Mobile
Computing
COMS 6998-11, Fall 2012
Instructor: Li Erran Li
([email protected])
http://www.cs.columbia.edu/~lierranli/coms
6998-11Fall2012/
12/11/2012: Course Summary
2
Syllabus
• Mobile App Development (lecture 2,3)
– Mobile operating systems: iOS and Android
– Development environments: Xcode, Eclipse with Android SDK
– Programming: Objective-C and android programming
• System Support for Mobile App Optimization (lecture 4,7)
– Mobile device power models, energy profiling and ebug debugging
– Core OS topics: virtualization, storage and OS support for power and context management
• Interaction with Cellular Networks (lecture 1,5, 8)
– Basics of 3G/LTE cellular networks
– Mobile application cellular radio resource usage profiling
– Measurement-based cellular network and traffic characterization
• Interaction with the Cloud (lecture 6,9)
– Mobile cloud computing platform services: push notification, iCloud and Google Cloud
Messaging
– Mobile cloud computing architecture and programming models
• Mobile Platform Security and Privacy (lecture 10,11,12)
– Mobile platform security: malware detection and characterization, attacks and defenses
– Mobile data and location privacy: attacks, monitoring tools and defenses
Cellular Networks and Mobile Computing
(COMS 6998-11)
3
Mobile App Development: iOS
•
•
•
•
•
•
•
iOS Overview
Objective C
Xcode
Model-View-Controller
Blocks and Multithreading
Core Data and Location
iCloud
Cellular Networks and Mobile Computing
(COMS 6998-11)
4
Mobile App Development: Android
• Android OS Overview
• Eclipse and Android SDK
• Application Framework
– Activity, content provider, broadcast receiver, intent
• Networking
• Google Cloud Messaging (GCM)
Cellular Networks and Mobile Computing
(COMS 6998-11)
System Support for Mobile App
Optimization
• Mobile device power models, energy profiler
and ebug debugging
• Core OS topics:
– Virtualization
– Storage
Cellular Networks and Mobile Computing
(COMS 6998-11)
5
System Calls As Power Triggers
Key observation: System call is the interface through which an
application communicates with the underlying system (hardware)
and outside world (Internet, GPS, etc.)
Key Idea: Use System Calls as triggers in power modeling
Advantages:
– Encapsulates utilization based triggers
• Parameters of system calls
– Captures power behavior of ones that do not
necessarily imply utilization
– Can be traced back to process, thread, function
• Eases energy accounting
Cellular Networks and Mobile Computing
(COMS 6998-11)
Courtesy: Pathak et al
6
Finite-State-Machine (FSM)
as Power Model Representation
7
Use Finite-State-Machine (FSM)
• Nodes: Power states
– Base State: No activity on phone
– Productive state: Actual utilization
– Tail state: No-useful work
• Edges: Transition rules
– System calls (start/completion)
– Workload (Ex: 50 pkts/sec)
– Timeout
Cellular Networks and Mobile Computing
(COMS 6998-11)
State
1
Transitions
State
3
Courtesy: Pathak et al
State
2
8
Virtualization: Device Namespace
safely,
correctly
multiplex
access to
devices
VP 1
VP 2
VP 3
•••
Cellular Networks and Mobile Computing
(COMS 6998-11)
•••
Android...
RTC / Alarms
Audio/Video
Sensors
Input
Power
Framebuffer
Cell Radio
WiFi
GPU
device namespaces
Linux
Kernel
Courtesy: Jason Nieh et al.
9
How Apps Use Storage?
• Exactly what makes web browsing slow on Android?
– Key lies in understanding how apps use SQLite and FS interface
/data/data/com.necla.webview
WebBench
Storage Schema
lib (empty)
cache
webviewCache
databases
These files written to
SQLite in sync
These files written to FS in
write-behind
6aaa3f00, 03051d8d, …
many files (5.5MB)
webview.db (14KB)
webviewCache.db (129KB)
 Apps typically store some data in FS (e.g., cache files)
and some in a SQLite database (e.g., cache map)
– All data through SQLite is written synchronously  slow!
– Apps often use SQLite oblivious to performance effects
Cellular Networks and Mobile Computing
(COMS 6998-11)
Courtesy: Nitin Agrawal et al.
10
Interaction with Cellular Networks
• Basics of 3G/LTE cellular networks
• Impact of radio access network on mobile apps
– Radio resource usage profiling (ARO)
• Impact of cellular network core on mobile
applications
– In-depth study of middleboxes in cellular networks
– Cellular network architecture characterization and
Implication to CDN
Cellular Networks and Mobile Computing
(COMS 6998-11)
11
LTE Infrastructure
•
•
•
eNodeB 1
•
Cellular Core Network
•
MME/PCRF/HSS
•
eNodeB 2
S-GW 1
UE 1
P-GW
eNodeB 3
•
S-GW 2
UE 2
GTP Tunnels
Cellular Networks and Mobile Computing
(COMS 6998-11)
UE: user equipment
eNodeB: base station
S-GW: serving
gateway
P-GW: packet data
network gateway
MME: mobility
management entity
HSS: home subscriber
server
PCRF: policy charging
and rule function
Internet and
Other IP Networks
12
LTE Architecture (Cont’d)
Control Plane
Data Plane
Mobility
Management
Entity
(MME)
User
Equipme
nt (UE)
Home
Subscriber
Server
(HSS)
•
eNodeB, S-GW and PGW are involved in
session setup, handoff,
routing
Policy Control and
Charging Rules
Function (PCRF)
Base
Serving
Station
(eNodeB)
Gateway
(S-GW)
Cellular Networks and Mobile Computing
(COMS 6998-11)
Packet Data
Network
Gateway
(P-GW)
13
Power Management: LTE
• UE runs radio resource
control (RRC) state
machine
• Two states: IDLE,
CONNECTED
• Discontinuous reception
(DRX): monitor one
subframe per DRX cylce;
receiver sleeps in other
subframes
Courtesy:Morley Mao
14
Power Management: UMTS
• State promotions have promotion delay
• State demotions incur tail times
Tail Time
Delay: 1.5s
Delay: 2s
Tail Time
Channel
Radio
Power
IDLE
Not
allocated
Almost
zero
CELL_FACH
Shared,
Low Speed
Low
CELL_DCH
Dedicated,
High Speed
High
Example: RRC State Machine
for a Large Commercial 3G Network
DCH Tail: 5 sec
Promo Delay: 2 Sec FACH Tail: 12 sec
Tail Time
Waiting inactivity timers to expire
DCH:
High Power State (high throughput and power consumption)
FACH: Low Power State (low throughput and power consumption)
IDLE: No radio resource allocated
Cellular Networks and Mobile Computing
(COMS 6998-11)
Courtesy: Feng Qian
15
16
ARO: Mobile Application Resource
Optimizer
• Motivations:
– Are developers aware of the RRC state machine and its
implications on radio resource / energy? NO.
– Do they need a tool for automatically profiling their prototype
applications? YES.
– If we provide that visibility, would developers optimize their
applications and reduce the network impact? Hopefully YES.
• ARO: Mobile Application Resource Optimizer
– Provide visibility of radio resource and energy utilization.
– Benchmark efficiencies of cellular radio resource and battery
life for a specific application
Cellular Networks and Mobile Computing
(COMS 6998-11)
Courtesy: Feng Qian et al.
RRC State Machine Inference
• State promotion inference
– Determine one of the two promotion procedures
– P1: IDLEFACHDCH;
P2:IDLEDCH
P1: IDLEFACH, P2:IDLEDCH
P1: FACHDCH, P2:Keep on DCH
Normal RTT < 300ms
RTT w/ Promo > 1500ms
A packet of min bytes never triggers FACHDCH promotion (we use 28B)
A packet of max bytes always triggers FACHDCH promotion (we use 1KB)
• State demotion and inactivity time inference
– See paper for details
Cellular Networks and Mobile Computing (COMS
6998-11)
17
18
ARO System Architecture
Cellular Networks and Mobile Computing
(COMS 6998-11)
Example: Pandora Music
Problem: High resource overhead of periodic audience measurements (every 1
min)
Recommendation: Delay transfers and batch them with delay-sensitive transfers
Cellular Networks and Mobile Computing
(COMS 6998-11)
Courtesy: Feng Qian
19
20
Impact of Middleboxes
IP spoofing creates security vulnerability
IP spoofing should be disabled
Small TCP timeout timers waste user device energy
Firewall Timer should be longer than 30 minutes
Out-of-order packet buffering hurts TCP performance
Consider interaction with application carefully
NAT
One NAT mapping linearly increases port # with time
Port prediction is feasible
Cellular Networks and Mobile Computing
(COMS 6998-11)
Impact of Architecture
• Observation
– All 4 major carriers cover the U.S. with only 4-8
clusters
– Cellular DNS resolvers are placed at the same level
as GGSN data centers
• Implication
– Mobile content providers should place their content
close to GGSNs
– Mobile content providers should select the content
server closest to the GGSN
Cellular Networks and Mobile Computing (COMS
6998-11)
21
22
Interaction with Cloud
• Mobile cloud platform services: push
notification, iCloud and Google Cloud
Messaging
• Mobile cloud computing architecture and
programming models
Cellular Networks and Mobile Computing
(COMS 6998-11)
Mobile Cloud Platform Services
• Social network services
– Demo: add social feature to the calculator app
• iCloud service
– Demo: add iCloud feature to the calculator app
• Push notification service
– Apple push notification service
• Demo: add push notification to the calculator app
– Google GCM
• Demo: add push notification to the calculator app
– Thialfi: reliable push notification system
• Track service
Cellular Networks and Mobile Computing (COMS
6998-11)
23
mCloud Programming Model
• MAUI: RPC based offloading architecture
• CloneCloud: tight synchronization between cloud and
phone
• Odessa: data-flow graph to exploit parallelism in
perception applications
• COMET: distributed shared memory
• MAUI, CloneCloud , Odessa all have profiler, solver
Remote
execution
unit
MAUI
CloneCloud
Odessa
COMET
Methods
(RMI)
Threads
(method
entry/exit)
Tasks
Threads
(any place)
Cellular Networks and Mobile Computing (COMS
6998-11)
24
Distributed Shared Memory

COMET is offloading + DSM



DSM usually applied to cluster environments


Offloading bridges computation disparity
DSM provides logically shared address space
Low latency, high throughput
Mobile relies on wireless communication
Cellular Networks and Mobile Computing (COMS
6998-11)
Courtesy: Mark Gordon et. al.
25
Java Memory Model


Dictates which writes a read can observe
Specifies 'happens-before' partial order



Access in single thread totally ordered
Lazy Release Consistency locking
Fundamental memory unit is the field

Known alignment, known width
Cellular Networks and Mobile Computing
(COMS 6998-11)
26
Field DSM


Track dirty fields locally
Need 'happens-before' established?


Transmit dirty fields! (mark fields clean)
Not clear it scales well past two endpoints


Not important to our motivation
Use classic cluster DSM on server
Cellular Networks and Mobile Computing
(COMS 6998-11)
27
VM-Synchronization



Used to establish ‘happen-before’ relation
Directed operation between pusher and
puller
Synchronizes



Bytecode sources
Java thread stacks
Java heap
Cellular Networks and Mobile Computing
(COMS 6998-11)
28
29
Mobile Security
• DroidRanger: Non-virtualization-based malware
detection
– Behavioral footprint matching for known malware
– Dynamic execution monitoring for unknown malware
• DroidScope Virtualization-based malware
detection
– Reconstruct OS, Dalvik VM and native view
• Malware characterization
–
–
–
–
Installation
Activation
Malicious payloads
Evolution
Cellular Networks and Mobile Computing
(COMS 6998-11)
30
Mobile Privacy
Data privacy
• Detecting and preventing privacy leaks
– PiOS for iOS
– TaintDroid for Android
• Limiting mobile data exposure with idle eviction [assume device
prone to loss]
– CleanOS
• Stealthy information leaks through covert channels and prevention
– Soundcomber
Location privacy [after-class reading]
– Quantifying location privacy
Cellular Networks and Mobile Computing
(COMS 6998-11)
31
PiOS: Analysis (CFG)
• Most iOS apps are written in Objective-C
• Cornerstone: objc_msgSend dispatch function
• Task: Resolve type of receiver and value of selector for
objc_msgSend calls
– Backwards slicing
– Forward propagation of constants and types
• Result: Inter and intra procedural CFG is constructed
from successfully resolved objc_msgSend calls
Cellular Networks and Mobile Computing
(COMS 6998-11)
32
PiOS: Finding Privacy Leaks
•
Inter and intra procedural Control Flow
Graph
•
Reachability Analysis (find paths)
– From interesting sources
– To network sinks
•
Implicit interruption of CFG for user-input
(e.g., dialog boxes, etc.)
– Touch events are generated by the OS not in the
developer's code
Cellular Networks and Mobile Computing
(COMS 6998-11)
Courtesy: Egele et. al
33
PiOS: Example ObjC to ASM
•
•
•
•
•
•
•
•
•
1 LDR
2 LDR
3 LDR
4 LDR
5 BLX
6 LDR
7 LDR
8 BLX
…
R0, =off_24C58
R1, =off_247F4
R0, [R0]
R1, [R1]
_objc_msgSend
R1, =off_247F0
R1, [R1]
_objc_msgSend
UIDevice
currentDevice
r0?
UIDevice
r1?
::currentDevice
uniqueIdentifier
r1?
UIDevice
::uniqueIdentifier
9 STR
10 LDR
R0, [SP,#0x60+var_34]
R3, [SP,#0x60+var_34]
…
11 BLX _objc_msgSend NSString ::initWithFormat:(fmt:
"uniqueid=%@&username=%@&country=%@&email=%@")
…
12 BLX _objc_msgSend POSTScore ::startPostingData:toURL: (0x1b478)
Cellular Networks and Mobile Computing
(COMS 6998-11)
34
TaintDroid
Leverage Android Platform Virtualization
Message-level tracking
Application
code
Virtual
machine
msg
Application
code
Virtual
machine
native system libraries
Network interface
Secondary storage
Cellular Networks and Mobile Computing
(COMS 6998-11)
Variable-level
tracking
Method-level
tracking
File-level
tracking
Courtesy: Byung-Gon et. al
TaintDroid Android Architecture in
Detail
35
36
The End
• Questions and comments?
Cellular Networks and Mobile Computing
(COMS 6998-11)