Deterministic Networking Lab Part Bernhard Frmel Institut fr - - PowerPoint PPT Presentation

Deterministic Networking Lab Part Frömel

Deterministic Networking Lab Part

Bernhard Frömel

Institut für Technische Informatik Technische Universität Wien

• 182.730 Deterministic Networking VU

SS14

• 23. 05. 2014

1/34

Deterministic Networking Lab Part Frömel

Motivation Emergence Self- Organization E versus SO

Part I

Emergence and Self-Organization

2/34

Deterministic Networking Lab Part Frömel

Motivation Emergence Self- Organization E versus SO

Fireflies synchronize

3/34

Deterministic Networking Lab Part Frömel

Motivation Emergence Self- Organization E versus SO

Fireflies synchronize

We ”understand” them1!

1http://web.eecs.utk.edu/~mclennan/Classes/

420-594-F07/NetLogo/Firefly.html

4/34

Deterministic Networking Lab Part Frömel

Motivation Emergence Self- Organization E versus SO

Flocking birds

◮ www.lalena.com/AI/Flock/Flock.aspx ◮ ”Emergent behavior in flocks” [1] 5/34

Deterministic Networking Lab Part Frömel

Motivation Emergence Self- Organization E versus SO

Internet

◮ World Wide Web: number of links high for few pages, low

for most pages2

◮ TCP based flows synchronize at network bottle necks,

simultaneous inc-/decrease of throughput

2http://internet-map.net/

6/34

Deterministic Networking Lab Part Frömel

Motivation Emergence Self- Organization E versus SO

Emergent Phenomena

◮ No generally accepted definition of emergence

◮ strong versus weak emergence ◮ show up as a surprise (subjectively perceived properties

useful?)

◮ ⇒ open research

◮ ’Sensible’ definition:

”Emergence: A phenomenon of a whole at the macro-level is emergent if and only if it is new with respect to the non-relational phenomena of any of its proper parts at the micro-level.” AMADEOS, Conceptual Model

7/34

Deterministic Networking Lab Part Frömel

Motivation Emergence Self- Organization E versus SO

Characteristics of Emergence [2]

◮ Emergent properties are:

◮ Interacting Parts: Parts need to interact, parallelism is not

enough

◮ Decentralized Control: only local mechanisms are used to

influence global behavior

◮ Coherence: logical and consistent correlation of parts at

micro-level ⇒ persistent pattern regardless of added/removed parts

◮ Micro-Macro effect: effect that comes into existence at

macro level (also called emergent) by interaction of parts at the microlevel

◮ Two-Way Link: emergent has causal effect on behavior of

parts at micro-level

◮ Radical Novelty: emergent not explicitly defined

◮ Origin:

◮ Non-linear behavior of parts ◮ Feedback/Feedforward mechanisms ◮ Time delays

8/34

Deterministic Networking Lab Part Frömel

Motivation Emergence Self- Organization E versus SO

Self-Organization

◮ Working definition:

”Self-Organisation is a dynamical and adaptive process where systems acquire and maintain structure themselves, without external control.” [2]

◮ Properties of Self-Organization:

◮ Autonomy: absence of external control ◮ Increase in Order: convergence to confined set in state

space

◮ Adaptability/Robustness: convergence robust w.r.t.

perturbation and changes

9/34

Deterministic Networking Lab Part Frömel

Motivation Emergence Self- Organization E versus SO

Emergence (E) versus Self-Organization (SO)

◮ Not synonyms! ◮ Both are dynamic processes arising over time ◮ E robust w.r.t. entering/leaving parts at micro-level ◮ SO robust w.r.t. changes of input and maintaining

increased order

◮ One without the other possible (see [2]) ◮ In combination able to structure complex systems by

keeping constituent parts simple

◮ Linking E and SO, different viewpoints:

◮ SO causes E: interaction of parts are SO, SO situated at

micro-level

◮ SO effect of E: emergents become more organized, SO is a

property of E

10/34

Deterministic Networking Lab Part Frömel

Clock Sync System Model Protocol

Part II

Distributed Clock Synchronization

11/34

Deterministic Networking Lab Part Frömel

Clock Sync System Model Protocol

Self-Stabilizing Distributed Clock Synchronization [3]

◮ Problem: synchronize all local clocks up to precision π

◮ achieve and maintain precision π across all independent

local clocks by exchange of messages

◮ no central control ◮ unknown initial conditions (i.e., local clock values arbitrary)

◮ How to do that? 12/34

Deterministic Networking Lab Part Frömel

Clock Sync System Model Protocol

Self-Stabilizing Distributed Clock Synchronization [3]

◮ Problem: synchronize all local clocks up to precision π

◮ achieve and maintain precision π across all independent

local clocks by exchange of messages

◮ no central control ◮ unknown initial conditions (i.e., local clock values arbitrary)

◮ How to do that? ◮ Solution: Emergence + Self-Organization

◮ Execute a protocol locally to achieve desired global effect ◮ Without external control input

12/34

Deterministic Networking Lab Part Frömel

Clock Sync System Model Protocol

Asynchronous Distributed System Model

◮ Nodes (processors) contain local oscillators with bounded

drift rate ρ, arbitrary phase

◮ Local oscillator generates clock ticks that are counted by

discrete LocalTimer

◮ Nodes interconnected by directed channels according to

topology (strongly connected, no self-loops, no multi-edges)

◮ Source node broadcasts messages to all directly

connected destination nodes

◮ Delivery order of messages arbitrary ◮ No-fault assumption: all nodes execute protocol correctly,

all communication channels transport messages reliably according to specified parameters

13/34

Deterministic Networking Lab Part Frömel

Clock Sync System Model Protocol

Drift Rate Bound ρ and Relative Drift δ(t)

◮ Drift of an oscillator is the frequency ratio of that oscillator

and a reference oscillator oscillating perfectly aligned to real-time

◮ Drift rate is

|driftosc − 1|

◮ Assumption: oscillators have a known bounded drift rate ρ:

0 < ρ << 1

◮ Maximum drift of fastest LocalTimer (discrete) over a time

duration t is:

(1 + ρ)t

◮ Maximum drift of slowest LocalTimer:

(1 + ρ)−1t

◮ Maximum relative drift δ(t):

δ(t) = ((1 + ρ) − (1 + ρ)−1)t.

14/34

Deterministic Networking Lab Part Frömel

Clock Sync System Model Protocol

Communication Delays

◮ Communication delay D, bounded: D ≥ 1 ◮ Network imprecision d, bounded: d ≥ 0 ◮ Communication latency γ:

γ = (D + d).

Figure : Event-Response Delay and Network Impression [3]

15/34

Deterministic Networking Lab Part Frömel

Clock Sync System Model Protocol

Protocol Description

◮ System has two states:

◮ synchronized: all nodes are within precision π ◮ unsynchronized: during start-up, dynamic changes of

nodes

◮ Synchronization protocol executed at each node

transitions system to synchronized state (convergence)

◮ Synchronization protocol must be repeatedly reexecuted

to maintain synchronized state (closure, stability)

◮ Nodes communicate by exchange of Sync messages ◮ Node times-out in case it’s LocalTimer reaches max. value

P (resynchronization period), LocalTimer resets

16/34

Deterministic Networking Lab Part Frömel

Clock Sync System Model Protocol

Protocol Execution

◮ Node restarts resynchronization process if

◮ LocalTimer times-out, or ◮ a Sync message is received

◮ Time-out ⇒ broadcast Sync message ◮ Received Sync message: ⇒ reset LocalTimer and relay

Sync message

◮ Eventually all nodes participate in (re)synchronization

process

◮ Prevent cascading effects: Ignore temporally close Sync

messages following a Sync message (ignore window)

17/34

Deterministic Networking Lab Part Frömel

Clock Sync System Model Protocol

Protocol in Pseudocode

Executed each time step:

18/34

Deterministic Networking Lab Part Frömel

TTEthernet Develop- ment Cluster TTEthernet Simulation Tools

Part III

Lab Environment

19/34

Deterministic Networking Lab Part Frömel

TTEthernet Develop- ment Cluster TTEthernet Simulation Tools

Development Environment

◮ Gbit/s TTEthernet Development System

◮ Four nodes (x86, Ubuntu 10.04 LTS, 2.6.32), redundant

TTEthernet switch setup

◮ Available Demo application showing video&audio

streaming (best-effort vs time-triggered)

◮ login: demonstrator / demo26 ◮ don’t update the whole distributions (installing additional

software via sudo apt-get install should be safe (in most cases (probably)))

◮ work on ’Video Client 4’

◮ All TTEthernet Tool DVDs/CDs: ~/Desktop/tte_cds 20/34

Deterministic Networking Lab Part Frömel

TTEthernet Develop- ment Cluster TTEthernet Simulation Tools

Building TTEthernet Applications [5]

◮ Define network configuration ◮ Implement application code ◮ Create the schedule (TTE Demo Scheduler) ⇒ *.xml ◮ Compile applications ◮ Create device configurations (TTE Build) ⇒ *.hex ◮ Load switches (TTE Load) ◮ Start applications 21/34

Deterministic Networking Lab Part Frömel

TTEthernet Develop- ment Cluster TTEthernet Simulation Tools

Building/Changing the Schedule

???

22/34

Deterministic Networking Lab Part Frömel

TTEthernet Develop- ment Cluster TTEthernet Simulation Tools

Omnet++, INET, and CoRE4INET [4]

◮ Omnet++ is an open-source network simulation framework

to build simulators

◮ wired, wireless, on-chip, queueing networks, ... ◮ Eclipse based IDE ◮ graphical visualization of simulation

◮ INET framework: an open-source communication networks

simulation package

◮ support for: UDP, TCP, IPv4, IPv6, Ethernet, 802.11, 802.1e

(QoS extension), 802.16 (WiMAX), ...

◮ CoRE4INET: extension of INET for real-time Ethernet

◮ TTEthernet (SAE AS6802), Time-Sensitive Networking

(TSN), formerly known as: IEEE 802.1 Audio Video Bridging (AVB))

◮ host-, switch-, and clock models ◮ host contains implementation of TTEthernet-API

23/34

Deterministic Networking Lab Part Frömel

TTEthernet Develop- ment Cluster TTEthernet Simulation Tools

CoRE4INET, INET Integration [4]

24/34

Deterministic Networking Lab Part Frömel

TTEthernet Develop- ment Cluster TTEthernet Simulation Tools

CoRE4INET, In Action

25/34

Deterministic Networking Lab Part Frömel

Tasks Logistics and Grading

Part IV

Assignment

26/34

Deterministic Networking Lab Part Frömel

Tasks Logistics and Grading

Tasks

• 1. Implement a distributed clock synchronization protocol

◮ Based on the paper:

” A Self-Stabilizing Distributed Clock Synchronization Protocol for Arbitrary Digraphs”, Mahyar R. Malekpour

◮ Use (real) TTEthernet for simple four nodes topology ◮ Use simulation framework for simulating hundreds of

nodes in different topologies (Schedule?)

• 2. Compare achieved clock precision π over time on

conventional Ethernet and TTEthernet (or other Time-Triggered Ethernets in simulation)

◮ Under different (self-chosen) network load/fault scenarios ◮ Conduct measurements, use TTEthernet global clock

• 3. Discuss results

’Hint’: The best-effort (traffic class) ’solution’ of the group from last year is available in the lab environment.

27/34

Deterministic Networking Lab Part Frömel

Tasks Logistics and Grading

Deliverables

◮ Implementation ◮ Documentation/Lab report

◮ English ◮ Include rudimentary HowTo develope TTEthernet

applications (tool usage + schedules)

◮ Focus on concise presentation of the implementation and

discussion of results

28/34

Deterministic Networking Lab Part Frömel

Tasks Logistics and Grading

Logistics and Grading

◮ Start: now ◮ Finish: September (latest) ◮ Work in groups, group size depends on number of

participants

◮ Location: Institute Lab ‘Fallstudienlabor’ ◮ Offer: weekly meetings ◮ Grading

◮ Deliverables: 75 points ◮ Delivery Talk/Presentation of results: 25 points

29/34

Deterministic Networking Lab Part Frömel

Summary Q&A Credits References

Part V

End

30/34

Deterministic Networking Lab Part Frömel

Summary Q&A Credits References

Summary

◮ Emergence and self-organization ◮ Self-Stabilizing distributed clock synchronization ◮ Available lab equipment and environment ◮ Assignment 31/34

Deterministic Networking Lab Part Frömel

Summary Q&A Credits References

Questions & Answers

32/34

Deterministic Networking Lab Part Frömel

Summary Q&A Credits References

Credits

◮ Images:

◮ https://www.flickr.com/photos/jamesjordan/ ◮ https://www.flickr.com/photos/87310153@N07/ ◮ https:

//www.flickr.com/photos/richardsmith155/

◮ https://www.flickr.com/photos/53297845@N06/ ◮ http://www.automationworld.com/sites/default/

files/styles/lightbox/public/field/image/ SynchClock.jpg?itok=DEFoWZod

33/34

Deterministic Networking Lab Part Frömel

Summary Q&A Credits References

References

[1] Felipe Cucker and Steve Smale. Emergent behavior in flocks. Automatic Control, IEEE Transactions on, 52(5):852–862, 2007. [2] Tom De Wolf and Tom Holvoet. Emergence versus self-organisation: Different concepts but promising when combined. In Engineering self-organising systems, pages 1–15. Springer, 2005. [3] Mahyar R Malekpour. A self-stabilizing distributed clock synchronization protocol for arbitrary digraphs. National Aeronautics and Space Administration, Langley Research Center, 2011. [4] Till Steinbach, Hermand Dieumo Kenfack, Franz Korf, and Thomas C. Schmidt. An Extension of the OMNeT++ INET Framework for Simulating Real-time Ethernet with High Accuracy. In SIMUTools 2011 – 4th International OMNeT++ Workshop, pages 375–382, New York, USA, March 21-25 2011. ACM DL. [5] TTTech. TTEthernet Introduction Workshop, Slides. TTTech Computertechnik AG, 2013.

34/34