This paper considers the use of lock-free shared objects within hard real-time systems. As the name suggests, lock-free shared objects are distinguished by the fact that they are not locked. As such, they do not give rise to priority inversions, a key advantage over conventional, lock-based object-sharing approaches. Despite this advantage, it is not immediately apparent that lock-free shared objects can be employed if tasks must adhere to strict timing constraints. In particular, lock-free object implementations permit concurrent operations to interfere with each other, and repeated interferences can cause a given operation to take an arbitrarily long time to complete. The main contribution of this paper is to show that such interferences can be bounded by judicious scheduling. This work pertains to periodic, hard real-time tasks that sharelock-free objects on a uniprocessor. In the first part of the paper, scheduling conditions are derived for such tasks, for both static and dynamic pri...

We show that previous algorithmic and scheduling work concerning the use of lock-free objects in hard real-time systems can be extended to support real-time transactions on memory-resident data. Using our approach, transactions are not susceptible to priority inversion or deadlock, do not require complicated mechanisms for data-logging or for rolling back aborted transactions, and are implemented as library routines that require no special kernel support. 1 Introduction In most real-time database systems, conventional mechanisms suchaslocks, timestamps, and serialization graphs are used for concurrency control. The main problem when using any of these mechanisms is that of handling con#icting operations. If an operation of a transaction creates a con#ict, then one of two strategies may be used: either that transaction may be blocked, or one or more of the transactions involved in the con- #ict may be aborted. When using con#ict resolution schemes that employ blocking, deadlockisakey ...

Mutual exclusion is a topic that Leslie Lamport has returned to many times throughout his career. This article, which is being written in celebration of Lamport's sixtieth birthday, is an attempt to survey some of his many contributions to research on this topic.

A space efficient wait-free algorithm for implementing a shared buffer for real-time multiprocessor systems is presented in this paper. The commonly used method to implement shared buffers in real-time systems is based on mutual exclusion. Mutual exclusion is penalised by blocking that typically leads to difficulties in guaranteeing deadlines in real-time systems. Researchers have introduced nonblocking algorithms and data structures that address the above problems. Many of the non-blocking algorithms that appeared in the literature have very high space demands though, some even unbounded, which makes them not suitable for real-time systems. In this paper we look at a simple, elegant and easy to implement algorithm that implements a shared buffer but uses unbounded time-stamps and we show how to bound the time-stamps by using the timing information that is available in many real-time systems. Our analysis and calculations show that the algorithm resulting from our approach is space efficient. The protocol presented here can support an arbitrary number of concurrent read and write operations.