5.5. Survivability of Large Distributed Systems
The purpose of decentralization in the previous discussion was to increase security and performance in the context of data storage systems. The usage of shares to decentralize data was used to increase security. The second property of the distribution schemes was the potential for increased performance, which for data storage means larger data bandwidth.
In this section, we want to address decentralization to increase performance in the context of distributed computing. In distributed computing, a large degree of distribution, or parallelism, is utilized to deal with large computational problems. For example, global computing systems like GRID or peer-to-environments use thousands of resources ...
Get Information Assurance now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
