Distributed Algorithms

Chapter 17 Shared Memory versus Networks

In the previous chapter, we described synchronizers, which comprise one method for simplifying the programming of asynchronous networks. This method enables (non-fault-tolerant) synchronous network algorithms such as those described in Chapter 4 to be used in asynchronous networks. In this chapter, we describe a second strategy for simplifying the programming of asynchronous networks: using them to simulate asynchronous shared memory systems. This enables asynchronous shared memory algorithms such as those described in Chapters 10, 11, and 13 to be used in asynchronous networks. Many other asynchronous shared memory algorithms can also be adapted to run in asynchronous networks, including practical algorithms ...