Real-Time Scheduling of Concurrent Transactions in Multidomain Ring Buses

Abstract

Abstract-We address the problem of scheduling concurrent periodic real-time transactions on Multi-Domain Ring Bus (MDRB). The problem is challenging because although the bus allows multiple non-overlapping transactions to be executed concurrently, the degree of concurrency depends on the topology of the bus and of executed transactions. This is a different challenge compared to that of multi-processor real-time task scheduling. To solve this problem, first, we propose two novel efficient scheduling algorithms for topographically-acyclic transaction sets. The first algorithm is optimal for transaction sets under restrictive assumptions while the second one induces a competitive sufficient schedulable utilization bound for more general transaction sets. Then, we extend these two algorithms for the scheduling of topographically-cyclic transaction sets. Extensive simulations show that the proposed algorithm can schedule transaction sets with high bus utilization and is better than that of related works in most practical settings. The implementation of the algorithms in a real test-bed shows that they have relatively low execution-time overhead.

Energy-Efficient Scheduling of Real-Time Tasks in Reconfigurable Homogeneous Multicore Platforms

“…We denote in the following by M i,j , the message to be exchanged between a pair of tasks τ i and τ j . According to [18] the communication model is based on: (i) the regular inter-arrival time T M i,j , (ii) the spent time to transmit a message W M i,j , (iii) the absolute deadline DM i,j , and (iv) the amount of exchanged data SM i,j . We note that period T M i,j of message M i,j is equal to period T i of sending task τ i (T M i,j =T i ).…”

“…According to the works reported in [18]- [20]- [22], to evaluate the communication feasibility between two cores, the related medium is considered as a virtual processor. Based on this hypothesis, the utilization of the communication medium between two cores C k and C l , denoted U Com (C k , C l ), is given by…”

Message scheduling for real-time interprocessor communication

“…A direct communication line will be established between sender and receiver of a message, which is a major advantage over existing networks-on-chip topologies, e.g. a ring bus topology as shown in [5] that has to store and forward data packets via routing nodes. Another valuable feature for real-time behavior is that the communication latency for any pair of processors does not depend on the distance between the processors.…”

A hardware architecture to deploy complex multiprocessor scheduling algorithms

“…The problem of task scheduling has several similarities with scheduling of network flows and bus transactions. To understand how algorithmic complexity affects overhead, the case of POGen [7] -a bus transactions scheduler -represents a good example. POGen uses slotted time and the notion of scheduling interval to perform scheduling: the scheduler is activated at the beginning of each scheduling interval and, each time, a table driven schedule for all the slots within current scheduling interval is computed.…”

“…For example, consider a system with m processors and n tasks: the most efficient implementation of Pfair [5], known as PD 2 [6], has a complexity of O(min(n, m * log n)). Similarly, POGen [7] -a scheduling algorithm for bus transactions -has a complexity of O(n 2 * h), with h being the length of hyperperiod for the considered n bus transactions.…”