Schedulability Driven Partitioning and Mapping for MultiCluster RealTime
Schedulability -Driven Partitioning and Mapping for Multi-Cluster Real-Time Systems Paul Pop, Petru Eles, Zebo Peng, Viaceslav Izosimov Embedded Systems Lab (ESLAB) Linköping University, Sweden 1 of 1/ 1415
Heterogeneous Network Distributed Heterogeneous System . . . No. Cs. . . Factory Systems Heterogeneous Networks Multi-Cluster Systems Automotive Electronics 2 of 2/ 1415
Distributed Safety-Critical Applicati. . . § Applications distributed over the heterogeneous networks Gateway . . . § Reduce costs: use resources efficiently § Requirements: close to sensors/actuators § Applications distributed over heterogeneous networks are difficult to. . . § Analyze(guaranteeing timing constraints) § Design(partitioning, mapping, bus access optimization) [DATE’ 03] This paper! 3 of 3/ 1415
Outline § Motivation è System architecture and application model § Scheduling for multi-clusters [DATE’ 03] § Design optimization problems § Partitioning § Mapping § Bus access optimization § Branch and bound optimization strategy § Experimental results § Contributions and Message 4 of 4/ 1415
Hardware Architectur. . . Time-triggeredcluster . . . Gateway § Static cyclic scheduling § Time-triggered protocol Event-triggeredcluster § Fixed priority preemptive scheduling § Controller area network protocol Time Triggered Protocol (TTP) § § Bus access scheme: time-division multiple-access (TDMA) Schedule table located in each TTP controller: message descriptor list (MEDL) S 0 S 1 Slot S 2 SG S 0 S 1 S 2 Controller Area Network (CAN) § § § Priority bus, collision avoidance Highest priority message wins the contention Priorities encoded in the frame identifier SG TDMA Round Cycle of two rounds 5 of 5/ 1415
. . . Software Architectur . . . Gateway Time-triggered cluster NG CPU CAN Controller N 1 CPU Out. CAN m 3 T T m 1 m 2 P 2 MBI TTP Controller Out. N 2 P 4 Out. TTP P 1 N 2 CPU CAN Controller m 3 MBI P 3 Event-triggered cluster SG S 1 Round 2 6 of 6/ 1415
Multi-Cluster Scheduling [DATE’ 0 Application, Partitioning, TT Bus Mapping, Configuration Architecture Priorities Static Scheduling Response Times Offsets Response Time Analysis Multi-Cluster Scheduling Schedule Tables Response times § Multi. Cluster. Scheduling algorithm § Schedulability analysis: communication delays through the gateway § Scheduling: cannot be addressed separately for each cluster 7 of 7/ 1415
Problem Formulatio § Input § System architecture § Application § Partialpartitioning and mapping, based on the designer’s experience § Output § Design implementation such that the application is schedulable . . . § Partitioning for each un-partitioned process Partitioning and. . . § Mapping for each un-mapped process mapping § Priorities for ET messages Communication Application : set of sequence process graphs Architecture : Multi-cluster infrastructure § TDMA slot and sizes for the TT bus § Priorities for ET processes Scheduling information § Schedule table for TT messages 8 of 8/ 1415
Motivational. Example#1 In which cluster to place process P 4? Deadline for P 5 N 1 (slower) N 3 Preemption not allowed P 1 (faster) N 2 P 4 P 2 Deadline for P 6 P 4 P 3 P 43 Met P 6 P 6 Met Preempted Missed P 4 P 5 PMet 4 P 5 Preempted P 1 P 2 P 3 P 4 P 5 P 6 N 1 70 X X X N 2 X X 50 70 X 40 N 3 X 40 X 90 40 X CAN TTC N 1 N 2 ETC N 3 TTP 9 of 9/ 1415
Motivational. Example#2 Where to map process P 2? Deadline N 3 N 2 CAN N 4 P 1 m 1 P 2 m 2 P 3 S 2 S 3 S 1 T SG m 2 S 23 S 2 S 3 S 1 SG T m 2 NG P 1 m 1 TTP P 2 m 1 N 1 Met Missed P 3 P 2 N 1 N 2 N 3 N 4 P 1 20 X X X P 2 X 40 X 50 P 3 X X 20 X CAN TTC N 1 N 2 N 3 NG ETC N 4 TTP 10 of 10/ 1415
Motivational. Example#3 What are the priorities on ETC? Which slot should come first on the TTC? Deadline Missed P 4 Round 4 P 1 Round 1 T T m 42 CAN N 2 P 3 m 1 P 1 m 2 P 2 m 3 P 4 m 4 SG 1 m S 13 m 3 S 1 m 4 SG 1 m 4 SG TTTT T m 2 m 1 m 4 NG SG 1 PP 2 2 PMet 4 Missed P 4 S 1 SG TT m 4 TTP SG 1 m 1 SGm 2 m SG 1 m 3 m 23 m 32 N 1 P 3 P 3 CAN P 1 P 2 P 3 P 4 N 1 20 X X 40 N 2 X 40 20 X TTC N 1 ETC N 2 TTP 11 of 11/ 1415
Optimization. Strategy § Multi-Cluster Configuration 1. Branch and Bound. Partitioning and Mapping § § § Branching rule Selection rule Bounding rule: lower bound 2. Bus Access Optimization § § § Determines the slot sequence and lengths on the TTC, message priorities on the ETC Greedy optimization heuristic Straightforward solution § § Partitioning and mapping that balances the utilization of processors and buses Could be produced by a designer without optimzation tools 12 of 12/ 1415
Experimental Result Percentage schedulable applications [%] Can we increase the number of schedulable applications? 100% 90% 80% 70% 60% 50% Multi-cluster Configuration 40% 30% 20% 10% 0% 50 100 150 200 Number of processes 250 Straightforward solution 13 of 13/ 1415
Experimental Results, Con How time-consuming is our optimization strategy? Average execution time [minutes] 45 Multi-Cluster Configuration 40 35 30 25 20 15 § Case study 10 5 0 50 100 150 200 Number of processes § Vehicle cruise controller § Distributed over the Straightforward TT and ET clusters solution 250 14 of 14/ 1415
Contributions and Messag § Contributions § Addressed design problems characteristic to multi-clusters § Partitioning § Mapping § Bus Access Optimization § Proposed a branch and boundapproach for optimization Analysis and optimization methods are needed for the efficient implementation of applications distributed over interconnected heterogeneous networks. 15 of 15/ 1415
- Slides: 15