Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

FRAP: A Flexible Resource Accessing Protocol for Multiprocessor Real-Time Systems

Shuai Zhao, Hanzhi Xu, Nan Chen, Ruoxian Su, Wanli Chang

TL;DR

This work tackles predictability and efficiency of resource sharing in fully-partitioned fixed-priority real-time multiprocessor systems by replacing rigid spin-priority rules with flexible per-resource spinning in FRAP. A novel minimum-cost maximum-flow (MCMF) based blocking analysis bounds the joint blocking $B_i+W_i$ under flexible spinning, ensuring predictable worst-case behavior, while a spin-priority assignment tunes $P_i^k$ to favor urgent tasks. Empirical evaluation shows FRAP yields substantial schedulability gains over MSRP, PWLP, MrsP, and Hybrid approaches (average improvements of 15.2–32.7%, up to 65.85%), with significantly lower analysis cost than ILP-based methods. The results demonstrate that fine-grained, urgency-aware spinning provides robust performance across diverse resource scenarios, enabling practical deployment in real-time systems.

Abstract

Fully-partitioned fixed-priority scheduling (FP-FPS) multiprocessor systems are widely found in real-time applications, where spin-based protocols are often deployed to manage the mutually exclusive access of shared resources. Unfortunately, existing approaches either enforce rigid spin priority rules for resource accessing or carry significant pessimism in the schedulability analysis, imposing substantial blocking time regardless of task execution urgency or resource over-provisioning. This paper proposes FRAP, a spin-based flexible resource accessing protocol for FP-FPS systems. A task under FRAP can spin at any priority within a range for accessing a resource, allowing flexible and fine-grained resource control with predictable worst-case behaviour. Under flexible spinning, we demonstrate that the existing analysis techniques can lead to incorrect timing bounds and present a novel MCMF (minimum cost maximum flow)-based blocking analysis, providing predictability guarantee for FRAP. A spin priority assignment is reported that fully exploits flexible spinning to reduce the blocking time of tasks with high urgency, enhancing the performance of FRAP. Experimental results show that FRAP outperforms the existing spin-based protocols in schedulability by 15.20%-32.73% on average, up to 65.85%.

FRAP: A Flexible Resource Accessing Protocol for Multiprocessor Real-Time Systems

TL;DR

This work tackles predictability and efficiency of resource sharing in fully-partitioned fixed-priority real-time multiprocessor systems by replacing rigid spin-priority rules with flexible per-resource spinning in FRAP. A novel minimum-cost maximum-flow (MCMF) based blocking analysis bounds the joint blocking under flexible spinning, ensuring predictable worst-case behavior, while a spin-priority assignment tunes to favor urgent tasks. Empirical evaluation shows FRAP yields substantial schedulability gains over MSRP, PWLP, MrsP, and Hybrid approaches (average improvements of 15.2–32.7%, up to 65.85%), with significantly lower analysis cost than ILP-based methods. The results demonstrate that fine-grained, urgency-aware spinning provides robust performance across diverse resource scenarios, enabling practical deployment in real-time systems.

Abstract

Fully-partitioned fixed-priority scheduling (FP-FPS) multiprocessor systems are widely found in real-time applications, where spin-based protocols are often deployed to manage the mutually exclusive access of shared resources. Unfortunately, existing approaches either enforce rigid spin priority rules for resource accessing or carry significant pessimism in the schedulability analysis, imposing substantial blocking time regardless of task execution urgency or resource over-provisioning. This paper proposes FRAP, a spin-based flexible resource accessing protocol for FP-FPS systems. A task under FRAP can spin at any priority within a range for accessing a resource, allowing flexible and fine-grained resource control with predictable worst-case behaviour. Under flexible spinning, we demonstrate that the existing analysis techniques can lead to incorrect timing bounds and present a novel MCMF (minimum cost maximum flow)-based blocking analysis, providing predictability guarantee for FRAP. A spin priority assignment is reported that fully exploits flexible spinning to reduce the blocking time of tasks with high urgency, enhancing the performance of FRAP. Experimental results show that FRAP outperforms the existing spin-based protocols in schedulability by 15.20%-32.73% on average, up to 65.85%.
Paper Structure (21 sections, 6 theorems, 11 equations, 5 figures, 6 tables, 2 algorithms)

This paper contains 21 sections, 6 theorems, 11 equations, 5 figures, 6 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. System Model
  3. Existing Resource Sharing Protocols
  4. Resource Sharing Protocols with Spin Locks
  5. Timing Bounds of FIFO Spin-based Protocols
  6. FRAP: Working Mechanism and Properties
  7. Blocking Analysis of FRAP
  8. The Problem of Bounding $B_i$ and $W_i$
  9. Identifying Sources of $B_i$ and $W_i$
  10. Bounding $B_i+W_i$ via MCMF
  11. Spin Priority Assignment of FRAP
  12. The Process of Spin Priority Assignment
  13. Implementing the Assignment
  14. Summary and Discussion
  15. Evaluation
  16. ...and 6 more sections

Key Result

Lemma 1

FRAP is compliant with the interference bound of FP-FPS system, i.e., $I_i=\sum_{\tau_h\in{\text{lhp}(i)}} \left\lceil{\frac{R_{i}}{T_h}}\right\rceil \cdot {\overline{C_h}}$.

Figures (5)

  • Figure 1: A system for illustrating the blocking effects.
  • Figure 2: Construction of $Q_2^k$ for $\tau_2$.
  • Figure 3: The problem of bounding $B_2$ and $W_2$ for $\tau_2$ in Tab. \ref{['tab:example_system']}.
  • Figure 4: An illustrative MCMF network for bounding $B_i+W_i$.
  • Figure 5: System schedulability with $N=5$, $M=12$, $A=5$, $L=\left [1 \mu s,100 \mu s \right ]$, $rsf=0.4$, and $K = M$ resources.

Theorems & Definitions (19)

  • Lemma 1
  • proof
  • Lemma 2
  • proof
  • Lemma 3
  • proof
  • Lemma 4
  • proof
  • Definition 1
  • Definition 2
  • ...and 9 more