problem 12: large distributed systems

Problem 12: Large Distributed Systems

ECU1

BUS

CC1

ECU2CC2

ECU3CC3

S1

S2

S3

S4

S5

5 Real-Time Input Streams- with jitter- with bursts- deadline > period

3 ECU’s with own CC’s

12 Tasks & 7 Messages-with different WCED-fully preemptive & independent

2 Scheduling Policies- Earliest Deadline First (ECU’s)- Fixed Priority (ECU’s & CC’s)

Hierarchical Scheduling- Static & Dynamic Polling Servers

Bus with TDMA- 4 time slots with different lengths (#1,#3 for CC1, #2 for CC3, #4 for CC3)

Total Utilization:- ECU1 59 %- ECU2 87 %- ECU3 67 %- BUS 56 %

Specification Data

ECU1 BUS(TDMA)

C1.1

C1.2

C2.1

C3.1

C4.1

C5.1

C3.2

T1.1

T1.3T2.1

T3.1

T3.3

PS

FPFP

CC1

ECU2

T4.1

T5.1

FP

CC2

ECU3

T1.2

FP FP

CC3

T3.2

FP

EDF

T2.2

PS

T4.2

PS

T5.2

S1

S2

S3

S4

S5

S1

S3

Within FP-blocks priorities are assigned from top (highest) to bottom (lowest).Communication tasks that share a TDMA slot use FP-scheduling.

Questions

• Is it schedulable?

• Minimum reqiured buffers?

• End-to-end delays of all streams?

Periodic Server

• A polling server can be thought of as a periodic task T (p, e). • When the server-task T is selected to run by the scheduler, the

server-task checks whether workload is waiting to be processed by the server.

• If yes, the server will provide e resources to process the waiting workload.

• If no work is available for the server, the task will immediately be finished, i.e. the server will not check for arriving work anymore until the next period starts.

• While the server is processing, it is preemptable. When it is preempted, it is put back into the ready-queue.

• If the server could not provide e resources during a time interval T, even though there was always enough work available, then this is treated the same as a deadline-miss. I.e. the system is considered to be not schedulable.