Pedro F. Souto

A group membership protocol for communication systems with both static and dynamic scheduling

We present a group membership protocol specially designed for next generation communication systems for real-time safety-critical applications such as FlexRay and FTT-CAN. The proposed protocol imposes an overhead of two bits per processor per communication cycle, when the system is in a quiescent state, and is able to tolerate benign failures of up to half of the group members between consecutive executions. Additionally, it removes a faulty processor within two communication cycles in the worst case and reintegrates a processor at the latest two communication cycles after it recovers. Compared with protocols developed for similar systems, it is as tolerant as the most robust protocol with a traffic overhead slightly higher than the most efficient protocol, which is much less robust.

A DHT-based approach for Path Selection and Message Forwarding in IEEE 802.11s industrial Wireless Mesh Networks

Wireless Mesh Networks (WMNs) are a promising communication technology that may offer greater flexibility and reliability, when compared to traditional wireless networks. WMNs open up new applications domains, but still need to find efficient mechanisms to deal with scalability and timeliness requirements. This paper proposes a scheme for Path Selection and Message Forwarding in IEEE 802.11s networks, that is suitable to be used in industrial environments. We present the DHT-based Cluster Routing Protocol (DCRP), a routing protocol based on DHTs, clustering of nodes and use of proxies. DCRP allows to improve the overall network performance by reducing the time required for path selection and the number of communication hops in large sized networks.

Unified overhead-aware schedulability analysis for slot-based task-splitting

Hard real- time multiprocessor scheduling has seen, in recent years, the flourishing of semi-partitioned scheduling algorithms. This category of scheduling schemes combines elements of partitioned and global scheduling for the purposes of achieving efficient utilization of the system’s processing resources with strong schedulability guarantees and with low dispatching overheads. The sub-class of slot-based “task-splitting” scheduling algorithms, in particular, offers very good trade-offs between schedulability guarantees (in the form of high utilization bounds) and the number of preemptions/migrations involved. However, so far there did not exist unified scheduling theory for such algorithms; each one was formulated in its own accompanying analysis. This article changes this fragmented landscape by formulating a more unified schedulability theory covering the two state-of-the-art slot-based semi-partitioned algorithms, S-EKG and NPS-F (both fixed job-priority based). This new theory is based on exact schedulability tests, thus also overcoming many sources of pessimism in existing analysis. In turn, since schedulability testing guides the task assignment under the schemes in consideration, we also formulate an improved task assignment procedure. As the other main contribution of this article, and as a response to the fact that many unrealistic assumptions, present in the original theory, tend to undermine the theoretical potential of such scheduling schemes, we identified and modelled into the new analysis all overheads incurred by the algorithms in consideration. The outcome is a new overhead-aware schedulability analysis that permits increased efficiency and reliability. The merits of this new theory are evaluated by an extensive set of experiments.

The Carousel-EDF scheduling algorithm for multiprocessor systems

We present Carousel-EDF, a new hierarchical scheduling algorithm for a system of identical processors, and its overhead-aware schedulability analysis based on demand bound functions. Carousel-EDF is an offshoot of NPS-F and preserves its utilization bounds, which are the highest among algorithms not based on a single dispatching queue and that have few preemptions. Furthermore, with respect to NPS-F, Carousel-EDF reduces by up to 50% the number of context switches and of preemptions caused by the high-level scheduler itself. The schedulability analysis we present in this paper is grounded on a prototype implementation of Carousel-EDF that uses a new implementation technique for the release of periodic tasks. This technique reduces the pessimism of the schedulability analysis presented and can be applied, with similar benefits, to other scheduling algorithms such as NPS-F.

Assessment of the Interference caused by uncontrolled traffic sources upon real-time communication in IEEE 802.11-based mesh networks

In this paper we assess the interference caused by external (uncontrolled) traffic sources upon real-time flows in IEEE 802.11-based mesh networks. Through a set of simulations, we assess the impact of external traffic sources by analyzing the end-to-end delay, frame loss ratio and deadline miss ratio as performance metrics for a set of real-time flows. We infer the maximum interfering traffic load that can be supported so that it does not negatively impact upon the behavior of the real-time flows. We highlight that there is a critical threshold load for the network above which the network can no longer reliably support real-time flows.

DHT-based Cluster Routing Protocol for IEEE802.11s Mesh networks

The IEEE 802.11s draft standard defines a mesh network in which frame delivery is done by forwarding the frame through nodes, called Mesh Points (MPs). To make this possible, it specifies two routing protocols: HWMP and RA-OLSR. Both protocols suffer from scalability issues caused by the use of broadcast messages for discovery and update of routes. In this work, we propose a new routing protocol, the DHT-based Cluster Routing Protocol (DCRP), that improves the scalability of 802.11s networks. Our approach is based on two mechanisms: clustering of nodes and DHT-based searching. Clustering allows to reduce the number of broadcast messages required for routing as well as the amount of routing information broadcasted. DHT-based searching is used to make up for the required routing information that is not diffused by the DCRP itself. Some back-of-the-envelope calculations indicate that our approach increases the scalability of the routing protocol.

Real-Time Communication in IEEE 802.11 Networks: Timing Analysis and a Ring Management Scheme for the VTP-CSMA Architecture

Keeping up with the timing constraints of real-time traffic in wireless environments is a hard task. One of the reasons is that the real-time stations have to share the same communication medium with stations that are out of the sphere-of control of the real-time architecture. That is, with stations that generate timing unconstrained traffic. The VTP-CSMA architecture targets this problem in IEEE 802.11 wireless networks. It is based on a Virtual Token Passing procedure (VTP) that circulates a virtual token among real-time stations, enabling the coexistence of real-time and non realtime stations in a shared communication environment. The worst-case timing analysis of the VTP-CSMA mechanism shows that the token rotation time is upper-bounded, even when the communication medium is shared with timing unconstrained stations. Additionally, the simulation analysis shows that the token rotation mechanism behaves adequately, even in the presence of error-prone communication channels. Therefore, the VTP-CSMA architecture enables the support of real-time communication in shared communication environments, without the need to control the timing behavior of every communicating device. A ring management procedure for the VTP-CSMA architecture is also proposed, allowing real-time stations to adequately join/leave the virtual ring. This ring management procedure is mandatory for dynamic operating scenarios, such as those found in VoIP applications.

A review of scalability and topological stability issues in IEEE 802.11s wireless mesh networks deployments

Scalability and topological stability are two of the most challenging issues in current wireless mesh networks WMNs deployments. In the literature, both the scalability and the topological stability of WMNs are described as likely to suffer from poor performance due to the ad hoc nature of the underlying IEEE 802.11 mechanisms. The main contribution of this article is a comprehensive review of the main topological stability and scalability-related issues in IEEE 802.11s-based networks. Moreover, the most relevant proposed solutions are surveyed, where both the drawbacks and the merits of each proposal are highlighted. At the end of the article, some open research challenges are presented and discussed. It is expected that this work may serve as motivation for more and deeper research on these issues to allow the design of future more stable and scalable IEEE 802.11s mesh networks deployments. Copyright

A forcing collision resolution approach able to prioritize traffic in CSMA-based networks

In this paper, a new Traffic Separation mechanism (TSm) is proposed for CSMA-based networks. The TSm mechanism is intended to be used as an underlying traffic separation mechanism, able to prioritize traffic in CSMA-based networks. It allows the coexistence of standard CSMA (non-modified) stations with TSm (modified) stations in the same communication domain. When a station implementing the TSm mechanism has a high-priority message to transfer, it will impose its transfer prior to any message from standard CSMA stations. This behavior guarantees the highest transmitting probability to the TSm-enabled station in open communication environments. Therefore, the TSm approach can be used as an underlying mechanism to build real-time communication systems upon CSMA-based networks. The behavior of the TSm mechanism was assessed by simulation in the case of a relevant CSMA-based network (IEEE 802.11). The simulation analysis shows that the TSm mechanism guarantees values for both the throughput and the average access delay that significantly improve the results obtained for standard IEEE 802.11 stations.

Mixed-criticality Scheduling with Dynamic Redistribution of Shared Cache

The design of mixed-criticality systems often involvespainful tradeoffs between safety guarantees and performance.However, the use of more detailed architectural modelsin the design and analysis of scheduling arrangements for mixedcriticalitysystems can provide greater confidence in the analysis,but also opportunities for better performance. Motivated by thisview, we propose an extension of Vestal 19s model for mixedcriticalitymulticore systems that (i) accounts for the per-taskpartitioning of the last-level cache and (ii) supports the dynamicreassignment, for better schedulability, of cache portions initiallyreserved for lower-criticality tasks to the higher-criticalitytasks, when the system switches to high-criticality mode. Tothis model, we apply partitioned EDF scheduling with Ekbergand Yi 19s deadline-scaling technique. Our schedulability analysisand scalefactor calculation is cognisant of the cache resourcesassigned to each task, by using WCET estimates that take intoaccount these resources. It is hence able to leverage the dynamicreconfiguration of the cache partitioning, at mode change, forbetter performance, in terms of provable schedulability. We alsopropose heuristics for partitioning the cache in low- and highcriticalitymode, that promote schedulability. Our experimentswith synthetic task sets, indicate tangible improvements inschedulability compared to a baseline cache-aware arrangementwhere there is no redistribution of cache resources from low- tohigh-criticality tasks in the event of a mode change.