Abstract

The Vestal model in widely used in the real-time scheduling community for representing mixed-criticality real-time workloads. This model requires that multiple WCET estimates -- one for each criticality level in a system -- be obtained for each task. Burns suggests that being required to obtain too many WCET estimates may place an undue burden on system developers, and proposes a simplification to the Vestal model that makes do with just two WCET estimates per task. Burns makes a convincing case in favor of adopting this simplified model; here, we report on our attempts at comparing the two models -- Vestal’s original model, and Burns’ simplification – with regards to expressiveness, as well as schedulability and the tractability of determining schedulability.

Meeting Name

3rd Workshop on Mixed-Criticality Systems, WMC (2015: Dec. 1, San Antonio, TX)

Department(s)

Computer Science

Comments

This research was supported in part by NSF grants CNS 1115284, CNS 1218693, CNS 1409175, and CPS 1446631, AFOSR grant FA9550-14-1-0161, ARO grant W911NF-14-1-0499, and a grant from General Motors Corp.

Document Type

Article - Conference proceedings

Document Version

Final Version

File Type

text

Language(s)

English

Rights

© 2015 The Authors, All rights reserved.

Publication Date

01 Dec 2015

Share

 
COinS