Optimized partitioning and priority assignment of real-time applications on heterogeneous platforms with hardware acceleration

With heterogeneous multi-core platforms being crucial to execute the highly demanding workloads of modern applications, memory-access predictability remains a key issue for the system's safety. Many solutions have been proposed over the years, but none has been applied on a large scale. Nowadays, we are in front of an unprecedented opportunity to have an impact on commercial platforms: the Memory System Resource Partitioning and Monitoring (MPAM) specification by Arm, which describes different memory-access regulation mechanisms, presenting a valuable industrial attempt to address this issue. However, several points of the specification are described at a high level only, leaving plenty of room for interpretation to hardware manufacturers. This paper takes a close look at the memory-access regulation mechanisms in the MPAM specification and provides some detailed instantiations of such mechanisms. A fine-grained memory contention analysis is presented for each of them to finally enable a comparison of their worst-case performance.

show abstract

“…2. The maximum term can be linearized using standard techniques that leverage auxiliary variables and the so-called big-M method [26].…”

Section: Baseline Analysis Frameworkmentioning

confidence: 99%

Analyzing Arm's MPAM From the Perspective of Time Predictability

Zini

Casini

Biondi

2023

IEEE Trans. Comput.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Finally, the inequality in Equation ( 9) sets the WCRT upper bound of τ i to be greater than or equal to the WCRT upper bound candidate corresponding to the point y i,g ∈ Y i for which E i,g = 1. This implementation of Equation ( 5) within the MILP formulation is analogous to the implementation presented in [32].…”

Section: Constraint 2 (Security-aware Wcrt) ∀τmentioning

confidence: 99%

Maximizing the Security Level of Real-Time Software While Preserving Temporal Constraints

et al. 2023

Self Cite

View full text Add to dashboard Cite

Embedded computing systems are becoming increasingly relevant in the Internet of Things (IoT) and edge computing domains, where they are often employed as the control entity of a cyber-physical system. When operating in such interconnected domains, a software system is susceptible to cyber-attacks from external agents, which can compromise the correct behavior of the system. In addition, the software executing in these systems is typically characterized by stringent timing constraints, which must be satisfied during system execution. Enabling software protections to enhance the security level of the embedded software comes at the cost of increasing the computation times of the tasks, introducing the risk of deadline misses that could also jeopardize the system behavior. This paper presents a methodology to optimize the security level of real-time software while preserving system-wide schedulability by leveraging timing analysis. The proposed approach is based on a mixed-integer linear programming (MILP) formulation that maximizes the security level of the tasks and integrates a response-time analysis technique to assess the schedulability of the system whenever additional protections are activated to shield the software against cyber-attacks targeting specific classes of vulnerabilities. An experimental evaluation is presented to assess the performance of the proposed approach on a representative set of tasks included in standard benchmarking suites for embedded software.

show abstract

“…Where hp(i) refers to the tasks with priorities higher or equal to the priority of the task T i An optimal deployment model DM * k is one that minimizes the Sum RT-ratio defined by [9] as a measure of performance for scheduling models, which is computed as follows:…”

Section: System Formalizationmentioning

confidence: 99%

“…In [8], the authors propose a new scheduling algorithm that maximizes the magnitude of safety margins while respecting the engineering constraints. In [9], new scheduling policies are proposed for heterogeneous platforms equipped with a hardware accelerator. When dealing with both problems (i.e., placement and scheduling), studies in the literature consider, in general, sub-optimal staged approaches.…”

Section: Introductionmentioning

confidence: 99%

PSRL: A New Method for Real-Time Task Placement and Scheduling Using Reinforcement Learning

Haouari,

Mzid,

Mosbahi

2023

International Conferences on Software Engineering and Knowledge Engineering

View full text Add to dashboard Cite

Modern real-time system development methodologies describe a stage in which application tasks are deployed onto an execution platform. The deployment process is divided into two steps: (i) task placement on processors and (ii) task scheduling to determine their execution order. The overall performance of the deployment model depends on the two steps, which are interdependent. In this paper, a new method based on reinforcement learning techniques, called PSRL, is proposed. PSRL explores all the feasible placements in the first step. In the second step, an optimal schedule is considered for each feasible placement. PSRL generates the optimal deployment, which corresponds to the placement and scheduling that minimize task response times. Application to case studies shows the applicability and quality of the obtained solutions when compared to related work.

show abstract

Optimized partitioning and priority assignment of real-time applications on heterogeneous platforms with hardware acceleration

Cited by 14 publications

References 30 publications

Analyzing Arm's MPAM From the Perspective of Time Predictability

Analyzing Arm's MPAM From the Perspective of Time Predictability

Maximizing the Security Level of Real-Time Software While Preserving Temporal Constraints

PSRL: A New Method for Real-Time Task Placement and Scheduling Using Reinforcement Learning

Contact Info

Product

Resources

About