Sinopsis
Parallel and distributed computing systems are now widely availabli . A parullel system consists of multiple processors that communicate with each otl er using shared memory. As the number of transistors on a chip increases, miiltipro essor chips will become fairly common. With enough parallelism available in applicE :ions, such systerns will easily beat sequential systems in performance. Figure 1.1 5 lows a parallel system with multiple processors. These processors communicate P ith each other using the shared memory. Each processor may also have local mem Iry that is not shared with other processors.
We define distributed systems as those computer syst.ems that co kain mult.iple processors connected by a communication network. In these syste 11s processors communicate with each other using messages that are sent over the I 2twork. Such systems are increasingly available because of decrease in prices of coa iuter processors and the high-bandwidth links t o connect them. Figure 1.2 shows t distributed system. The communication network in the figure could be a local ,rea network such as an Ethernet, or a wide area network such as the Internet.
Programming parallel and distributed systems requires a different set of tools and techniques than that required by the traditional sequential software. The focus of this book is on these techniques,.
Content
- Introduction
- Mutual Exclusion Problem
- Synchronization Primitives
- Consistency Conditions
- Wait-Free Synchronization
- Distributed Programming
- Models and Clocks
- Resource Allocation
- Global Snapshot
- Global Properties
- Detecting Termination and Deadlocks
- Message Ordering
- Leader Election
- Synchronizers
- Agreement
- Transactions
- Recovery
- Self-stabilization
,+A+-+Stroustrup,+Bjarne.jpg)
.jpg){kind=spacer}
0 komentar:
Posting Komentar