1. A Tool for Describing and Evaluating Hierarchical Real-Time Bus Scheduling Policies Author: Trevor Meyerowitz, Claudio Pinello, Alberto DAC2003, June 24,2003 Presenter : Tsung-Yu Ho (Seminar Book Page 81)

2. Abstract  We present a tool suite for building, simulating, and analyzing the results of hierarchical descriptions of the scheduling policy for modules sharing a bus in real-time applications. These schedules can be based on a variety of factors including characteristics of messages and time slicing and are represented in a hierarchical tree-like structure that specifies multiple levels of arbitration. This structure can describe many popular arbitration schemes. Our simulator evaluates the specified scheduling structure on a set of message traces for a given bus. We illustrate our approach by applying it to two examples: the SAE Automotive Benchmark and Voice Over IP (VoIP). Although this paper deals with just bus scheduling policies, the approach can be easily extended to other real-time scheduling problems.

3. Outline  What’s the Problem  Introduction Scheduling policies Propose a tool suite  Related work Various scheduling  Scheduler representation and evaluation Define representation Describe in language Evaluation  Experiment Result SAE Benchmark VoIP Benchmark

4. What’s the problem  Increased complexity of interaction between blocks  Previous work focus on selecting process mapping and communication topology  Ignore arbitration policy of bus.  In real-time system, most applications have real-time constrained, it is hard to analyze.

5. Introduction  Focus on the representation and evaluation of various scheduling policies  Real-time message among modules communicating via a shared bus.  Propose a tool suite (called STRANG)  Building, simulating, and analyzing the result of scheduling policy  Provide a simple hierarchical language Describe the arbitration policy Simulate the policy Easy to explore the design space

6. Scheduling Policies (Related work)  This paper is based on well-known scheduling  Focus on communicating scheduling (via shared bus)  Introduce the popular scheduling as follows: 1.Event-Triggered Scheduling 2.Time-Triggered Scheduling 3.Hybrid Scheduling

7. 1. Event Triggered Scheduling  Based on priority  FIFO ordering : simple to implement  Fixed Priority : simple to implement  EDA (Earliest Deadline First) : best result Dynamically give priority to message  CAN (Control Area Network) bus  Successfully used in event triggered scheduling  Use a fixed priority arbitration scheme Based on id message (id : identifier) Each node can use bus when there is no message being transmitted.  Has more flexible than Time-triggered scheduling

8. 2. Time-Trigger Scheduling  TDMA (Time Division Multiple Access) Policy  Divide a period time into several time slices  Each time slice is assigned to only one node Easy to ensure fairness between the nodes  TTP (Time-Triggered Protocol) bus  Use the TDMA policy A lower arbitration overhead than CAN bus Higher bandwidth utilization  Easy to have latencies Because of non periodic message  Inflexible than CAN bus

9. 3. Hybrid Scheduling  Improve Performance by combining Event- Triggered and Time-Triggered Scheduling  Provide the flexibility of CAN with determinism of TTP  Hybrid approaches usually is used in Multimedia domain (like VoIP)

10. Outline  What’s the Problem  Introduction Scheduling policies Propose a tool suite  Related work Various scheduling  Scheduler representation and evaluation Define representation Describe in language Evaluation  Experiment Result SAE Benchmark VoIP Benchmark

11. Definition  A hierarchical arbitration policy schedule message between p entities communicating via a shared bus  P : primary node  M : Message  Z B : Scheduling Policy  M B: a trace on the bus  Goal  Pick the best policy under fitness metric

12. Characteristics  Message and Nodes  Each message has its own priority Sender-id, receiver-id, size, message-id, arrival-time deadlines of message, time until deadline.  Primary Nodes choose available message to bus  Bus scheduling policy determine which message go first  Metrics  Find the quality metric The number of missed deadlines The overall execution time The average throughput of the bus  Then choose fitness policy

13. Tree Representation  Arbitration Syntax P : first custom operation A : arbitration node S : primary/sender node PolicyID: policy name used by node Alloc : the style of time allocation used by node Preemption : preemption policy

14. Sample Trees  CAN Tree  Event-Triggered  TTP Tree  Time-Triggered  Hybrid Tree  Combined two of these

15. Examples of TTP

16. Experiment Result (SAE Benchmark)  SAE (Society of Automotive Engineers)  Different protocol at bus speed 100Kbps, 125Kbps, and 250Kbps  53 message type between 7 node  5 second of message trace

17. SAE Benchmark Evaluation Result (1/1)

18. SAE Benchmark Evaluation Result(2/2)

19. Voice over IP Benchmark  G.729A voice codec  10 byte samples every 10 ms  Evaluate 4 types of arbitration policies  EDF, FIFO  Fixed Priority with RMS (Rate Monotonic Scheduling) non-preemptive, shortest periods have the highest priorities.  Fixed Priority with DMS (Deadline Monotonic Scheduling) shortest deadlines get the highest priorities

20. Conclusion  Formulate the problem of scheduling real- time messages on a shared bus  Show the benefit of using hierarchical arbitration policies for optimizing the schedule  Present a tool that can represent a wide variety of trees, and simulate them using message traces.  The results could easily be generalized to evaluate scheduling policies in a variety of other domains.