Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Pioneering Deterministic Scheduling and Network Structure Optimization for Time-Critical Computing Tasks in Industrial IoT

Yujiao Hu, Yining Zhu, Huayu Zhang, Yan Pan, Qingmin Jia, Renchao Xie, Gang Yang, F. Richard Yu

TL;DR

The paper tackles the problem of providing deterministic end-to-end response guarantees for periodic time-critical IIoT tasks by jointly optimizing computation and network resources. It introduces RTAP and NSOP formulations, proves their complexity, and derives resource-sharing conflict conditions. The proposed IIoTBroker delivers high-quality, low-overhead, deterministic scheduling across computation and network layers, while IIoTDeployer offers cost-aware network upgrades to satisfy task requirements with minimal spending. Together, they enable scalable, predictable IIoT operation in industrial environments and demonstrate favorable performance across diverse scenarios and cost configurations.

Abstract

The Industrial Internet of Things (IIoT) has become a critical technology to accelerate the process of digital and intelligent transformation of industries. As the cooperative relationship between smart devices in IIoT becomes more complex, getting deterministic responses of IIoT periodic time-critical computing tasks becomes a crucial and nontrivial problem. However, few current works in cloud/edge/fog computing focus on this problem. This paper is a pioneer to explore the deterministic scheduling and network structural optimization problems for IIoT periodic time-critical computing tasks. We first formulate the two problems and derive theorems to help quickly identify computation and network resource sharing conflicts. Based on this, we propose a deterministic scheduling algorithm, \textit{IIoTBroker}, which realizes deterministic response for each IIoT task by optimizing the fine-grained computation and network resources allocations, and a network optimization algorithm, \textit{IIoTDeployer}, providing a cost-effective structural upgrade solution for existing IIoT networks. Our methods are illustrated to be cost-friendly, scalable, and deterministic response guaranteed with low computation cost from our simulation results.

Pioneering Deterministic Scheduling and Network Structure Optimization for Time-Critical Computing Tasks in Industrial IoT

TL;DR

The paper tackles the problem of providing deterministic end-to-end response guarantees for periodic time-critical IIoT tasks by jointly optimizing computation and network resources. It introduces RTAP and NSOP formulations, proves their complexity, and derives resource-sharing conflict conditions. The proposed IIoTBroker delivers high-quality, low-overhead, deterministic scheduling across computation and network layers, while IIoTDeployer offers cost-aware network upgrades to satisfy task requirements with minimal spending. Together, they enable scalable, predictable IIoT operation in industrial environments and demonstrate favorable performance across diverse scenarios and cost configurations.

Abstract

The Industrial Internet of Things (IIoT) has become a critical technology to accelerate the process of digital and intelligent transformation of industries. As the cooperative relationship between smart devices in IIoT becomes more complex, getting deterministic responses of IIoT periodic time-critical computing tasks becomes a crucial and nontrivial problem. However, few current works in cloud/edge/fog computing focus on this problem. This paper is a pioneer to explore the deterministic scheduling and network structural optimization problems for IIoT periodic time-critical computing tasks. We first formulate the two problems and derive theorems to help quickly identify computation and network resource sharing conflicts. Based on this, we propose a deterministic scheduling algorithm, \textit{IIoTBroker}, which realizes deterministic response for each IIoT task by optimizing the fine-grained computation and network resources allocations, and a network optimization algorithm, \textit{IIoTDeployer}, providing a cost-effective structural upgrade solution for existing IIoT networks. Our methods are illustrated to be cost-friendly, scalable, and deterministic response guaranteed with low computation cost from our simulation results.
Paper Structure (38 sections, 5 theorems, 6 equations, 11 figures, 5 tables, 6 algorithms)

This paper contains 38 sections, 5 theorems, 6 equations, 11 figures, 5 tables, 6 algorithms.

Table of Contents

  1. Introduction
  2. Related Work
  3. Cloud/Edge/Fog-based IIoT Computation Offloading
  4. Deterministic Scheduling of TSN traffics
  5. Preliminaries
  6. TSN and IEEE 802.1Qbv
  7. Computing Resource Sharing
  8. The Lifecycle of IIoT Computing Tasks
  9. Theoretical Modeling
  10. RTAP Formulation
  11. NSOP Formulation
  12. Theoretical Analysis on Resource Sharing Conflicts
  13. Analysis on sharing computing resources
  14. Analysis on sharing network Links
  15. IIoTBroker
  16. ...and 23 more sections

Key Result

Theorem 1

If there are integers $m$ and $n$ such that $A \cap B \neq \emptyset$, where $A = [nP_{h_j}, (n+1)P_{h_j})$ and $B = [o_{h_i} + mP_{h_i}, o_{h_i} + \alpha_{h_i} + mP_{h_i})$, make $V = [ \max(o_{h_i} + mP_{h_i}-nP_{h_j}, 0), \min(P_{h_j}, o_{h_i} + \alpha_{h_i} + mP_{h_i}-nP_{h_j}))$. If $o_{h_j} \

Figures (11)

  • Figure 1: Illustration of how the IEEE 802.1Qbv TAS works on egress hardware. TSN traffic can be classified into three types: time-triggered traffic (TTT), audio video bridging traffic (AVBT), and best effort traffic (BET). With TAS, each queue is controlled by the gate control list that determines which queue is open. 1 indicates that the queue is open, while 0 indicates the queue is closed.
  • Figure 2: An example of an IIoT periodic time-critical computing task's lifecycle in one period. (a) visualizes the important life stages of the IIoT task $h_j$. (b) describes the lifecycles of $h_j$ from a temporal perspective.
  • Figure 3: An example of computing resource sharing conflicts of task $h_i$ and $h_j$. Blue and green blocks refer to the occupied time slots by $h_i$ and $h_j$, respectively.
  • Figure 4: Situation enumeration when $h_i$ and $h_j$ cannot share computing resource.
  • Figure 5: An illustration of the processes of LCFU. In this case, seven IIoT computing tasks are generated, and four routers and one server locate in the original IIoT network. (a) Build the ideal graph$G$. (b) IIoTBroker schedules $h_A$ and $h_B$ to $C_1$. (c) IIoTBroker schedules $h_C$ to virtual server $C_1$. (d) The connection between $C_2$ and $R_b$ is retained, connections between $C_2$ and other routers are removed from $G$. $C_2$ becomes a real server. (e) IIoTBroker schedules $h_D$ to server $C_2$. (f) IIoTBroker schedules $h_E$ to virtual server $C_6$. (g) The connection between $C_6$ and $R_c$ is retained, and connections between $C_6$ and other routers are removed from $G$. $C_6$ becomes a real server. IIoTBroker schedules $h_F$ to server $C_6$. (h) IIoTBroker schedules $h_G$ to the virtual server $C_7$. (i) The connection between $C_7$ and $R_d$ is retained, connections between $C_7$ and other routers are removed from $G$. $C_6$ becomes a real server. Other unused servers $C_3$, $C_4$, $C_5$ and related virtual connections are removed from $G$.
  • ...and 6 more figures

Theorems & Definitions (8)

  • Theorem 1
  • proof
  • Theorem 2
  • proof
  • Theorem 3
  • Theorem 4
  • Lemma 1
  • proof