Dependable End-to-End Delay Constraints for Real-Time Systems using - - PowerPoint PPT Presentation

dependable end to end delay constraints for real time
SMART_READER_LITE
LIVE PREVIEW

Dependable End-to-End Delay Constraints for Real-Time Systems using - - PowerPoint PPT Presentation

Dependable End-to-End Delay Constraints for Real-Time Systems using SDN Rakesh Kumar, Monowar Hasan, Smruti Padhy, Konstantin Evchenko, Lavanya Piramanayagam, Si Sibin Mo Mohan and Rakesh B. Bobba The 15th International Workshop on Real-Time


slide-1
SLIDE 1

Dependable End-to-End Delay Constraints for Real-Time Systems using SDN

Rakesh Kumar, Monowar Hasan, Smruti Padhy, Konstantin Evchenko, Lavanya Piramanayagam, Si Sibin Mo Mohan and Rakesh B. Bobba

The 15th International Workshop on Real-Time Networks June 27, 2017

slide-2
SLIDE 2

Overview

§ Time-critical real-time applications require:

  • A guaranteed u

upper b bound o

  • n t

the en end-to to-end p packet de delay

  • Avionics, automobiles, industrial control systems, power control networks, etc.

§ Current approach: Separate networks for different classes of traffic (high, medium, low criticality)

  • Higher costs
  • Increased management overheads: routers/switches have to be individually programmed
  • Increased attack surfaces

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

2

Ju June 27, 2017

slide-3
SLIDE 3

Software Defined Networking (SDN)

§ Logically centralized co cont ntrol l pla lane ne at co cont ntrolle ller § Standardized da data plane in commoditized sw switche hes and switch-controller communication protocol § Controller’s No Northbound d API

  • Enables find-grained control of individual flows in the network

3

Co Contr troller Switch Switch Switch Switch Switch

Ju June 27, 2017

Applications on Northbound API

slide-4
SLIDE 4

SDN Switch

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

4

Ju June 27, 2017

Port

slide-5
SLIDE 5

SDN Switch

§ Each switch port contains multiple queues

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

5

Port

Ju June 27, 2017

slide-6
SLIDE 6

SDN Switch

§ Each switch port contains multiple queues § The entire switch has a meter table

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

6

Port

Ju June 27, 2017

slide-7
SLIDE 7

SDN Switch

§ Each switch port contains multiple queues § The entire switch has a meter table § Flow Tables: Contain matching rules and options to select port, queue and meters

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

7

Port

Ju June 27, 2017

slide-8
SLIDE 8

Can SDN Help in Real-Time Systems?

§ SDN offers no no end end-to to-end end t tim iming ing g gua uarant ntees ees for packet flows of individual applications § SDN and real-time:

  • Can the SDN architecture enable computation of flow paths that meet real-time guarantees?

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

8

Ju June 27, 2017

slide-9
SLIDE 9

Problem Overview

§ Each flow (fk) with ba bandwidth th (Bk) and given end end-to to-end end de delay (Dk) requirements § Pr Problem: allocate n such flows so that the delay and bandwidth constraints are satisfied

  • For all f

flows

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

9

SCADA Controller Ethernet Relay

fk fk

Ju June 27, 2017

Overvi view/In Intuition è Separate Q Queue for E Each H High P Priority/Critical F Flow

slide-10
SLIDE 10

Motivating Example

§ Two switch, four host topology § Two simultaneous flows with different traffic send rates

  • Two different queue configuration:

1. Each flow has a separate q queue configured at 50 Mbps 2. Both flows share same q queue configured at 100 Mbps

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

10 10

Ju June 27, 2017

slide-11
SLIDE 11

Motivating Example

§ Two switch, four host topology § Two simultaneous flows with different traffic send rates

  • Two different queue configuration:

1. Each flow has a separate q queue configured at 50 Mbps 2. Both flows share same q queue configured at 100 Mbps

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

11 11

The c case w with s separate q queues ex experiences lower a average p per-pa packet de delay due t to l lack o

  • f i

f interference

Ju June 27, 2017

slide-12
SLIDE 12

Can SDN Help in Real-Time Systems?

§ SDN offers no no end end-to to-end end t tim iming ing g gua uarant ntees ees for packet flows of individual applications § SDN and real-time:

  • Can the SDN architecture enable computation of flow paths that meet real-time guarantees?

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

12 12

Ju June 27, 2017

YE YES

slide-13
SLIDE 13

Solution Approach

  • 1. Setup one flow at a time
  • Flows priorities are assigned in de

delay-mo monotonic order (tighter delay è higher priority)

  • 2. Access system state using the northbound API of the controller
  • E.g.: available resources, network topology
  • 3. Compute the flow path through the SDN such that its requirements are met
  • Solve as a multi-constraint path selection problem
  • 4. Realize path in the SDN topology by using the northbound API

13 13

Ju June 27, 2017

slide-14
SLIDE 14

Solution Approach

14 14

Admission C Control

Ju June 27, 2017

slide-15
SLIDE 15

Solution Approach (contd.)

§ End-to-end delay for a given flow can be composed from individual delays at nodes/links:

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

15 15

Ju June 27, 2017

slide-16
SLIDE 16

Solution Approach (contd.)

§ End-to-end delay for a given flow can be composed from individual delays at nodes/links:

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

16 16

SCADA Controller Ethernet Relay

fk

PATH (unknown!) Delay of a link

Ju June 27, 2017

slide-17
SLIDE 17

Solution Approach (contd.)

§ End-to-end delay for a given flow can be composed from individual delays at nodes/links: § Bandwidth utilization of the flow on the entire path:

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

17 17

SCADA Controller Ethernet Relay

fk

PATH (unknown!) Delay of a link Required bandwidth utilization of a link

Ju June 27, 2017

slide-18
SLIDE 18

Solution Approach (contd.)

Multi-Constraint Path (MCP) Selection

§ Delay constraint

  • Total delay over path less than end-to-end delay budget

§ Bandwidth constraint

  • Flow bandwidth utilization on all links can fit within the total utilization along the path

§ Shortest-path may NO NOT satisfy both the constraints!

  • MCP is NP-Complete!
  • Developed a

a p polynomial h heuristic to solve this multi-constraint problem è calculate paths

18 18

SCADA Controller Ethernet Relay

fk

PATH (unknown!)

Ju June 27, 2017

slide-19
SLIDE 19

Solution Approach (Contd.)

Path Realization Using Intents

§ Int Intent ent è actions performed on the packets in a given flow at an individual switch § Each intent is 4-tuple given by § Intents are realized with a fl flow rule and a corresponding ex exclusive qu queues

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

19 19

Ju June 27, 2017

slide-20
SLIDE 20

Evaluation

Setup

§ Experiments performed on a machine running Mi Mininet and RY RYU

  • Python implementation of northbound application for QoS Synthesis

§ 250 250 random topologies: five switches, each switch having two hosts § Each link has the bandwidth of 10 10 Mb Mbps § Link delays: generated uniformly randomly between [25, 25, 125 125] microseconds § Bandwidth requirements: randomly generated between [1, 1, 5] Mbps § [1, 1, 5] real-time and [1, 1,3] non-real time flows are generated using Ne Netperf

  • Each flow lasts for 10 seconds

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

20 20

Ju June 27, 2017

slide-21
SLIDE 21

Results

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

21 21

X-axis: D : Delay r requirements Y-axis: N : Number o

  • f fl

f flows Z-axis: % : % o

  • f

f sc schedul ulable fl flows

Ju June 27, 2017

slide-22
SLIDE 22

Results

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

22 22

X-axis: D : Delay r requirements Y-axis: N : Number o

  • f fl

f flows Z-axis: % : % o

  • f

f sc schedul ulable fl flows

The acceptance ratio de decreases with: 1. Increasing the number of flows; or 2. For stringent end-to-end delay requirements

Ju June 27, 2017

slide-23
SLIDE 23

Results

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

23 23

X-axis: N : Number o

  • f fl

f flows Y-axis: O : Observed d delay ( (99th

th pe

percentile)

Ju June 27, 2017

slide-24
SLIDE 24

Results

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

24 24

X-axis: N : Number o

  • f fl

f flows Y-axis: O : Observed d delay ( (99th

th pe

percentile)

  • 1. Non-real time flows do n

not c cause interference for real-time flows

  • 2. Increasing the number of real-time

flows increases end-to-end delay

Ju June 27, 2017

slide-25
SLIDE 25

Conclusion

§ Our approach:

  • Successfully allocate flows for highly critical RTS network traffic on SDN architectures
  • Non-critical flows do not interfere with critical ones
  • Useful for COTS systems

§ The evaluation results are another instance of the “No Free Lunch Theorem”

  • The acceptance ratio decreases either
  • With increasing the number of flows or
  • Stringent end-to-end delay requirements

§ Open Issues

  • What does the optimal allocation look like?
  • Multiplexing the usage of a single queue for multiple flows remains an open problem

RT RTN 2017 : Dependable End-to to-En End Delay Constraints For Real-Ti Time Sys ystems Using SDN

25 25

Ju June 27, 2017

slide-26
SLIDE 26

Thank You!

Questions?

June 27, 2017 RTN 2017 : Dependable End-to-End Delay Constraints For Real-Time Systems Using SDN

26 26

RTSS Prepint https://arxiv.org/abs/1703.01641