Pushing the limits of CAN - Scheduling frames with offsets provides a major performance boost
Upcoming SlideShare
Loading in...5
×
 

Pushing the limits of CAN - Scheduling frames with offsets provides a major performance boost

on

  • 447 views

M. Grenier, L. Havet, N. Navet, "Pushing the limits of CAN - Scheduling frames with offsets provides a major performance boost", Proc. of the 4th European Congress Embedded Real Time Software (ERTS ...

M. Grenier, L. Havet, N. Navet, "Pushing the limits of CAN - Scheduling frames with offsets provides a major performance boost", Proc. of the 4th European Congress Embedded Real Time Software (ERTS 2008), Toulouse, France, January 29 - February 1, 2008.

Statistics

Views

Total Views
447
Views on SlideShare
447
Embed Views
0

Actions

Likes
0
Downloads
5
Comments
0

0 Embeds 0

No embeds

Accessibility

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

    Pushing the limits of CAN - Scheduling frames with offsets provides a major performance boost Pushing the limits of CAN - Scheduling frames with offsets provides a major performance boost Presentation Transcript

    • Pushing the limits of CAN - Scheduling frames with offsetsprovides a major performance boost Nicolas NAVET INRIA / RealTime-at-Work http://www.loria.fr/~nnavet http://www.realtime-at-work.com Nicolas.Navet@loria.fr ERTS – 30/01/2008 Joint work with Mathieu GRENIER and Lionel HAVET
    • In-vehicle networking :will CAN be able to keep up the pace?Typically max. bus load is set to 35%Not enough wrt to short/medium term bandwidthneeds … Solution 1: multiple CAN networks … but gateways induce heavy overhead Solution 2: switch to FlexRay … expensive for bandwidth alone Solution 3: optimize the scheduling of CAN frame .. Offsets provide a solution to make CAN predictable at higher network load (≥60%) 2
    • Scheduling frames with offsets ?! Principle: desynchronize transmissions to avoid load peaks0 10 15 Periods0 5 5 20 ms0 0 5 15 ms 10 ms0 10 20 30 40 50 60 70 80 90 100 110 5 52,5 2,5 Periods 0 0 20 ms 15 ms 10 ms0 10 20 30 40 50 60 70 80 90 100 110 Algorithms to decide offsets are based on arithmetical properties of the periods and size of the frame 3
    • System model (1/2) Frame response time taskECU Frame Transmission request Higher prio. frames frameCAN Performance metric: worst-case response time 4
    • System model (2/2)The offset of a message stream is the time at whichthe transmission request of the first frame is issuedComplexity: best choosing the offsets is exponential inthe task periods → approximate solutionsMiddleware task imposes a certain granularityWithout ECU synchronisation, offsets are local to ECUs 5
    • But task scheduling has to be adapted… Frame response time taskECU Frame Transmission request Higher prio. frames frameCAN In addition, avoiding consecutive frame constructionson an ECU allows to reduce latency 6
    • Offsets Algorithm (1/3)Ideas: assign offsets in the order of the transmission frequencies release of the first frame is as far as possible from adjacent frames identify “least loaded interval”Ex: f1=(T1=10), f2=(T2=20), f3(T3=20)Time 0 2 4 6 8 10 12 14 16 18Frame f1,1 f2,1 f1,2 f3,1 7
    • Offsets Algorithm applied on a typical body network 65 ms 21 ms 8
    • Offsets Algorithm (3/3)Low complexity and efficient as is butfurther improvements possible: add frame(s) / ECU(s) to an existing design user defined criteria : optimize last 10 frames, a specific frame, take into account priorities optimization algorithms: tabu search, hill climbing, genetic algorithms … 9
    • Efficiency of offsets : some insight (1/2) Work = time totransmitthe CAN frames sent by the stations Almost a straight line, suggests that our algorithm is near-optimal 10
    • Efficiency of offsets : some insight (2/2) A larger workload waiting for transmission implieslarger response times for the low priority frames .. 11
    • Computing worst-caseresponse times with offsets
    • Computing frame worst-case response time with offsets AUTOSAR COM 5ms Frame-packing task Waiting queue:Requirements : -FIFO 2 -Highest Priority First- handle 100+ frames (HPF - Autosar) 1 -Carmaker specific- very fast execution times- ≠ waiting queue policy at the CAN Controllermicrocontroller level 9 6 8- limited number of transmissionbuffers buffer Tx CAN Bus 13
    • WCRT : State of the artScientific literature: Complexity is exponential No schedulability analysis with offsets in the distributed non-preemptive case Offsets in the preemptive case : not suited for > 10-20 tasks WCRT without offsets: infinite number of Tx buffers and no queue at the microcontroller levelOur software: NETCAR-Analyzer 14
    • NETCAR-Analyzer : developed at INRIA, then RealTime-at-Work 15
    • NETCAR-Analyzer : an overview Worst-case response time on CAN with and withoutoffsets Proven near-optimal offsets assignments with user-defined performance criteria (e.g. WCRT of the 10 lowest prio.frames) Exhibit the situations leading to the worst-case (results canbe checked by simulations/testing) Enable to dimension transmission/reception buffers(RAM) Handle both FIFO and prioritized ECUs Fast multi-core implementation (<1mn for 100 frames) Industrial use since December 2006 16
    • Performance evaluation : Experimental Setup WCRT of the frames wrt random offsetsand lower bound WCRT reduction ratio for chassis and bodynetworks Load increase : add new ECUs / add moretraffic
    • Experimental Setup Body and chassis networksNetwork #ECUs #Messages Bandwidth Frame periods Body 15-20 ≈ 70 125Kbit/s 50ms-2sChassis 5-15 ≈ 60 500Kbit/s 10ms-1sWith / without load concentration: one ECU generates 30%of the load Set of frames generated with NETCARBENCH (GPL-licenced) 18
    • Offsets in practice : large response time improvements (1/2) 65 ms 32 21 17 19
    • WCRT Reduction Ratio Body Networks Chassis NetworksResults are even better with loaded stations 20
    • Offsets allow higher network loads Typically: WCRT at 60% with offsets ≈ WCRT at 30% without offsets 21
    • Partial offset usage 65 ms 42 34 17 22
    • Conclusions Offsets provide an cost-effective short-term solution to postpone multiple CANs and FlexRay Tradeoff between Event and Time Triggered ET CAN CAN with offsets TT-CAN + Complexity + Determinism Further large improvements are possible bysynchronizing the ECUs … 23
    • Questions, feedback?please contact me atNicolas.Navet@loria.fr 24