Size-Based Queuing An Approach to Improve Bandwidth Utilization in - - PowerPoint PPT Presentation

▶
size based queuing
SMART_READER_LITE
LIVE PREVIEW

Size-Based Queuing An Approach to Improve Bandwidth Utilization in - - PowerPoint PPT Presentation

Jan 02, 2024 827 likes •1.24k views

Size-Based Queuing An Approach to Improve Bandwidth Utilization in TSN Networks Florian Heilmann and Gerhard Fohler RTN18 - Barcelona, Spain In this talk Highest-priority (HP) messages may experience jitter due to blocking by

slide-1
SLIDE 1

Size-Based Queuing

An Approach to Improve Bandwidth Utilization in TSN Networks Florian Heilmann and Gerhard Fohler RTN’18 - Barcelona, Spain

slide-2
SLIDE 2

In this talk

  • Highest-priority (HP) messages may experience jitter due to blocking

by lower-priority (LP) messages

  • In a TAS-enabled non-preemptive TSN network
  • Guard bands
  • used to avoid this jitter
  • But also decrease bandwidth utilization
  • Our approach to reclaim wasted bandwidth
  • Provide an initial analysis of our approach

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 2

slide-3
SLIDE 3

TSN and TAS primer

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 3

slide-4
SLIDE 4

Ethernet and Time-Sensitive Networking

  • Switched Ethernet is replacing existing network technologies
  • Cheap, mature technology
  • However: No timing/delivery guarantees

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 4

slide-5
SLIDE 5

Ethernet and Time-Sensitive Networking

  • Continuation of AVB TG
  • Goal is to make Ethernet safety-critical
  • Automotive, Avionics, Factory automation
  • Extend AVB, e.g.
  • higher precision clock sync
  • Frame preemption
  • Frame replication for redundancy

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 5

slide-6
SLIDE 6

Time-Aware Shaper and Guard Bands

  • Time-Aware Shaper (TAS) is part of TSN
  • TAS adds …
  • Gating mechanism for switch queues
  • Gate schedule to control the gates
  • TAS enables …
  • Scheduled transmissions
  • Ways to provide exclusive network access to a queue

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 6

slide-7
SLIDE 7

Time-Aware Shaper and Guard Bands

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 7

Ingress Port 1 Ingress Port 2 Ingress Port n Egress Port

Input queuing function 𝑅7 𝑅6 𝑅5 𝑅4 𝑅0 TX selection function

𝐻7 𝐻6 𝐻5 𝐻4 𝐻0

… … …

802.1Q Ethernet Switch

Safety-critical, highest-priority traffic (denoted HP in this presentation) Safety-critical, lower-priority traffic (denoted LP in this presentation) Best Effort traffic (not considered in this presentation)

slide-8
SLIDE 8

Time-Aware Shaper and Guard Bands

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 8

𝐻𝐼𝑄 𝐻𝑀𝑄

  • c

“Open” “Closed ” “Closed” “Open” 𝑢 𝑢 𝑢

𝑯𝑰𝑸

  • c

𝑯𝑴𝑸 𝒑𝒗𝒖

slide-9
SLIDE 9

Time-Aware Shaper and Guard Bands

  • Highest priority (HP) messages may experience blocking by

lower priority (LP) messages Unless:

  • Network provides message preemption functionality
  • Not considered in this work
  • Gate schedules are modified to avoid blocking
  • Guard bands (GB): Preemptively close gate of LP messages to ensure

idle network once HP gate opens

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 9

slide-10
SLIDE 10

Time-Aware Shaper and Guard Bands

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 10

𝑢 𝑢 𝑢

  • c

𝐻𝐼𝑄 𝐻𝑀𝑄

  • c

𝑝𝑣𝑢

HP Jitter

slide-11
SLIDE 11

Time-Aware Shaper and Guard Bands

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 11

𝑢 𝑢

  • c

𝐻𝐼𝑄 𝐻𝑀𝑄

𝑢

  • c

max

𝑝𝑣𝑢

Guardband

slide-12
SLIDE 12

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 12

𝑢 𝑢 𝑢

  • c

𝐻𝐼𝑄 𝐻𝑀𝑄 𝑝𝑣𝑢

  • c

Guardband

Time-Aware Shaper and Guard Bands

slide-13
SLIDE 13

Problem description

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 13

slide-14
SLIDE 14

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 14

𝑢 𝑢 𝑢

  • c

𝐻𝐼𝑄 𝐻𝑀𝑄

  • c

𝑝𝑣𝑢

Bandwidth Wastage caused by Guard Bands

Guardband

  • Guard bands cause time intervals where all

gates are closed

  • Pending messages can’t be transmitted.
  • Even though output link is idle

Output bandwidth utilization decreases (bandwidth is wasted)

slide-15
SLIDE 15

Approaches to reduce Jitter of HP messages

  • Guard Bands
  • Can result in reduced bandwidth utilization
  • Because we always need to consider largest message
  • Our approach: Size Based Queuing (SBQ)
  • Builds on the guard band approach

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 15

slide-16
SLIDE 16

Our approach: Size-Based Queuing

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 16

slide-17
SLIDE 17

Size-Based Queuing

  • Non-preemptive network +

Low Jitter requirement of HP messages  guard bands unavoidable

  • Can we modify existing guard bands?

Idea

  • Guard band size  Size of largest message in queue
  • Distribute messages into multiple queues according to their size
  • Control the size of the largest message in each queue

Shrink guard bands Improve bandwidth utilization

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 17

slide-18
SLIDE 18

Size-Based Queuing

  • Method overview
  • 1. Split some/all LP queues into queue-sets
  • 2. Define size-thresholds
  • 3. Modify 𝑔

𝑗𝑜 to distribute messages according to thresholds

  • 4. Add new gates for the new queues
  • 5. Update gate schedule or modify 𝑔

𝑝𝑣𝑢 to handle new queues

  • 6. Shrink guard bands where applicable

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 18

slide-19
SLIDE 19

SBQ Implementation Example

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 19

Ingress Port 1 Ingress Port 2 Ingress Port n Egress Port

Input queuing function 𝑅7 𝑅4 𝑅0 TX selection function

𝐻7 𝐻4 𝐻0

… … …

𝑅6

𝐻5

TSN TAS Ethernet Switch

𝑅5

𝐻6

slide-20
SLIDE 20

𝑅6

𝐻6 𝐻5

𝑅5 𝑅6 𝑅5

𝐻6 𝐻5

SBQ Implementation Example

𝑅5,𝑡𝑛

𝐻5,𝑡𝑛

𝑅5,𝑚𝑠𝑕

𝐻5,𝑚𝑠𝑕

𝑅6,𝑡𝑛

𝐻6,𝑡𝑛

𝑅6,𝑚𝑠𝑕

𝐻6,𝑚𝑠𝑕

𝜐6 𝜐5

TX selection function Input queuing function

slide-21
SLIDE 21

SBQ Implementation Example

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 21

Ingress Port 1 Ingress Port 2 Ingress Port n Egress Port

Input queuing function 𝑅7 𝑅4 𝑅0 TX selection function

𝐻7 𝐻4 𝐻0

… … …

𝑅5,𝑡𝑛

𝐻5,𝑡𝑛

𝑅5,𝑚𝑠𝑕

𝐻5,𝑚𝑠𝑕

𝑅6,𝑡𝑛

𝐻6,𝑡𝑛

𝑅6,𝑚𝑠𝑕

𝐻6,𝑚𝑠𝑕

𝜐6 𝜐5

SBQ Ethernet Switch

slide-22
SLIDE 22

Analysis

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 22

slide-23
SLIDE 23

Analysis - Bandwidth Utilization

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 23

𝑢 𝑢 𝑢

Guardband

  • c

𝐻𝐼𝑄 𝐻𝑀𝑄

  • c

𝑝𝑣𝑢

slide-24
SLIDE 24

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 24

𝑡𝑗𝑨𝑓𝑛𝑡𝑕 ≥ 𝜐𝑀𝑄 → 𝑅𝑀𝑄,𝑚𝑠𝑕 𝑡𝑗𝑨𝑓𝑛𝑡𝑕 < 𝜐𝑀𝑄 → 𝑅𝑀𝑄,𝑡𝑛

Analysis - Bandwidth Utilization

𝜐𝑀𝑄

slide-25
SLIDE 25

Analysis - Bandwidth Utilization

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 25

𝑢

𝐻𝐼𝑄

𝑢

  • c

𝑢

  • c

𝐻𝑀𝑄,𝑚𝑠𝑕

Guardband

𝑢

  • c

𝐻𝑀𝑄,𝑡𝑛

“Closed”

max

Guardband GB

𝑝𝑣𝑢

“Open”

slide-26
SLIDE 26

Analysis – Design Time Overheads

  • Queues added by SBQ have to be considered during network

design

  • Generate schedules that provide service to all queues
  • Make use of TSN shapers (CBS, BLS, etc.)
  • Use additional algorithms to arbitrate within queue-sets
  • E.g. Round-Robin

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 26

slide-27
SLIDE 27

Analysis - Latency

  • Changing the way messages are queued affects their latency

through a switch

  • Only LP messages affected
  • HP unaffected
  • (we don’t care about BE)

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 27

slide-28
SLIDE 28

Analysis - Latency

  • Individual queues of a queue set get only a portion of the

bandwidth

Time before message reaches front of queue is longer  Latency increases .

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 28

1

slide-29
SLIDE 29

Analysis - Latency

  • Messages are distributed into queue-set, message has to

compete against less other messages inside a queue

Message reaches front of the queue faster  Latency decreases .

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 29

2 1

slide-30
SLIDE 30

Analysis - Latency

  • (depending on arbitration scheme)
  • Arbitration within queue-sets affects latency of all LP messages

Latency of lower relative priority messages increases Latency of higher relative priority messages decreases

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 30

2 1 33

slide-31
SLIDE 31

Analysis - Latency

  • Actual number depend on SBQ parameters
  • Number of queues in a queue-set
  • Values of thresholds
  • Schedule/intra queue-set arbitration (algorithm, priorities)
  • Impact differs for each queue of the queue-set

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 31

1

2

2 1 2 1 33 3 3 33

slide-32
SLIDE 32

Analysis - Latency

However

  • Average latency of all messages is improved
  • Because of reclaimed bandwidth

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 32

1

2

2 1 2 1 33 3 3 33

slide-33
SLIDE 33

Analysis – Hardware Overheads

  • SBQ introduces additional HW requirements
  • Increased Memory requirements
  • for additional queues
  • Increased complexity
  • Input queuing function
  • Transmission (TX) selection function

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 33

slide-34
SLIDE 34

Analysis - Summary

  • Result in better bandwidth utilization
  • Lowers average latency of messages
  • Without affecting HP message transmissions
  • At the cost of hardware/design time overheads and latency of some
  • f the LP messages
  • Plug and Play with legacy TSN devices
  • Flexible (w.r.t. which priorities, ports, switches)

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 34

slide-35
SLIDE 35

Conclusion and Future Work

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 35

slide-36
SLIDE 36

Conclusion

  • TAS enabled TSN networks without preemption may incur

lowered bandwidth utilization caused by guard bands

  • SBQ can reclaim some of this wasted bandwidth and thus

improve average latency

  • Exact impact and cost TBD

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 36

slide-37
SLIDE 37

Future Work

  • Experimental evaluation / simulation of SBQ
  • How does each SBQ parameter affects latency/bandwidth
  • Evaluate different approaches w.r.t. 𝑔

𝑝𝑣𝑢 (RR, SP, etc.)

  • Check which other metrics of messages we can make use of

(P802.1Qcr)

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 37

slide-38
SLIDE 38

Thank you! … Questions?

slide-39
SLIDE 39

Backup – Message Overtaking

  • Message overtaking can occur with SBQ
  • E.g. small messages overtake large messages
  • Also present in e.g. IEEE 802.1CB (FRER)
  • Should be fixed on higher OSI layers

Or:

  • Keep message size of a flow fixed

Size-Based Queuing - An Approach to Improve Bandwidth Utilization in TSN Networks 42