Moore's Law: Fast, Cheap Computing and What It Means for the Manager
Moore's Law, named for the co-founder of Intel Gordon Moore, defines expected advances in the need for data storage over time. In reality, it defines much more, beyond simply data storage. Read this chapter and attempt the exercises to gain a broader understanding of the importance and costs associated with Information Systems.
Bringing Brains Together: Supercomputing and Grid Computing
- Most modern supercomputers use massive sets of microprocessors working in parallel.
- The microprocessors used in most modern supercomputers are often the same commodity chips that can be found in conventional PCs and servers.
- Moore's Law means that businesses as diverse as financial services firms, industrial manufacturers, consumer goods firms, and film studios can now afford access to supercomputers.
- Grid computing software uses existing computer hardware to work together and mimic a massively parallel supercomputer. Using existing hardware for a grid can save a firm the millions of dollars it might otherwise cost to buy a conventional supercomputer, further bringing massive computing capabilities to organizations that would otherwise never benefit from this kind of power.
- Massively parallel computing also enables the vast server farms that power online businesses like Google and Facebook, and which create new computing models, like software as a service (SaaS) and cloud computing.
- The characteristics of problems best suited for solving via multicore systems, parallel supercomputers, or grid computers are those that can be divided up so that multiple calculating components can simultaneously work on a portion of the problem. Problems that are linear - where one part must be solved before moving to the next and the next - may have difficulty benefiting from these kinds of "divide and conquer" computing. Fortunately many problems such as financial risk modeling, animation, manufacturing simulation, and gene analysis are all suited for parallel systems.