Source: Wikipedia. Pages: 47. Chapters: Locality of reference, Software performance testing, Algorithmic efficiency, Program optimization, Copy-on-write, Java performance, Memoization, Lookup table, Performance engineering, Profiling, Inline function, List of performance analysis tools, Object pool pattern, Run-time algorithm specialisation, String interning, Instruction path length, Code bloat, Object code optimizer, Hot spot, Browser speed test, MILEPOST GCC, Collective Tuning Initiative, Zero-copy, Memory footprint, Inner loop, Performance prediction, Instrumentation, Collective Optimization Database, Capacity optimization, Optimum programming, Performance Application Programming Interface, Deep sampling, Fast path. Excerpt: In computer science, efficiency is used to describe properties of an algorithm relating to how much of various types of resources it consumes. Algorithmic efficiency can be thought of as analogous to engineering productivity for a repeating or continuous process, where the goal is to reduce resource consumption, including time to completion, to some acceptable, optimal level. The two most frequently encountered and measurable metrics of an algorithm are:- but also might apply to and perhaps even (An extreme example of these metrics might be to consider their values in relation to a repeated simple algorithm for calculating and storing (p+n) to 50 decimal places running for say, 24 hours, on a processor operating in its own purpose-built heated or air conditioned unit.) The process of making code more efficient is known as optimization and in the case of automatic optimization (i.e. compiler optimization - performed by compilers on request or by default), usually focus on space at the cost of speed, or vice versa. There are also quite simple programming techniques and 'avoidance strategies' that can actually improve both at the same time, usually irrespective of hardware, software or language. Even the re-ordering of nested conditional statements - to put the least frequently occurring condition first (example: test patients for blood type ='AB-', before testing age > 18, since this type of blood occurs in only about 1 in 100 of the population - thereby eliminating the second test at runtime in 99% of instances), can reduce actual instruction path length, something an optimizing compiler would almost certainly not be aware of - but which a programmer can research relatively easily even without specialist medical knowledge. The first machines that were capable of computation were severely limited by purely mechanical considerations. As later electronic machines were developed they were, in turn, limited by the speed of their electronic counterparts. As software rep