To date, genetic algorithms have been shown to be a useful method for identifying optimal solutions for a wide range of scientific and statistical purposes. The applications and overall effect of genetic algorithms is therefore expected to continue to expand into the foreseeable future (Anderson-Cook, 2005)
273). The analysis of the performance-price ratio between parallel and sequential processing therefore requires an assessment of what benefits can be realized by integrating numerous separate processors compared to the costs that are involved in building these capacities into a single machine (Benkler, 2005)
Constraint: the content of each bin does not exceed L. (Chao, Harper & Quong, 1993)
, 2001). Researchers have found that genetic algorithms are especially useful for identifying solutions to optimization problems involving state spaces that exceed the abilities of stochastic dynamic programming approaches (Clark & Mangel, 2000)
One-dimensional bin packing represents a longstanding problem in computing that has a number of practical applications that are associated with the need to minimize space and/or time that make the use of genetic algorithms particularly useful. By any measure, the bin packing problem is challenging and complex and a wide range of heuristic solutions have been proposed over the years with the same overarching objective to pack a collection of objects into the minimum number of fixed-size "bins" (Coffman, Garey & Johnson, 1984)
The 1D-bin packing problem, though, does have some solutions available. For instance, the problem can be solved in a straightforward fashion through the use of a specific genetic algorithm known as the "grouping genetic algorithm," or GGA (Falkenauer, 2008)
The 1D-bin packing problem, though, does have some solutions available. For instance, the problem can be solved in a straightforward fashion through the use of a specific genetic algorithm known as the "grouping genetic algorithm," or GGA (Falkenauer, 2008)
Therefore, even the optimal processing speeds attainable in the future will remain constrained in this regard, but there are some alternative approaches to computer processing that can further increase the functionality of computers, including parallel computing and genetic algorithms which are discussed further below. Parallel Computing In computing, the term "parallelism" is used to describe a system's architecture, in other words, "The organization and interconnection of components of computer systems" (Faulkner, Senker & Velho, 1999, p
Sequential and Parallel Computing Simply stated, parallel computers consist of a group of processors that operate in a cooperative fashion in order to solve computational problems. This basic definition is sufficiently broad to describe even parallel supercomputers that are comprised of hundreds or even thousands of processors, as well as networks of workstations, multiple-processor workstations, and embedded systems (Foster, 2011)
138). This method, though, has some drawbacks since the master must await feedback from slaves, and like the weakest link in a chain, these algorithms only perform as fast as the slowest node (Haupt & Haupt, 2004)
223). The fundamental principles of genetic algorithms can be depicted through the use of a pseudo code as follows: begin initialize population evaluate initial population while not (stop condition) do begin select organisms select a genetic operator generate new population by applying a genetic operator evaluate newly generated population end Although all genetic algorithms will be unique in some fashion, the foregoing structure is a common attribute of all such algorithms (Joh et al
1). In principle, benchmarking analysis of very-large simulations typically involves relatively straightforward calculations that are applied to various components that are located in sparsely populated arrays that are intended to simulate enormous numbers of participants; at present, a three-tiered architecture is being considered to partition the benchmarking problem into parallel executing graphs structures in which one kernel will process linking sets of related members while concomitantly another kernel will process parallel searches for the most relevant results from these graphs with a final kernel remaining under development (Johnson, 2010)
Based on this determination, Faulknauer (1999) proposed a standard GGA methodology. In the years since this seminal work, there have been a number of applications of these concepts to various grouping problems, with differing levels of success, including the equal piles problem, graph coloring, edge coloring and the exam-timetabling problem (Lewis & Paechter, 2002)
37). Genetic algorithms have attracted a great deal of interest from software engineers and researchers based on their ability to identify designs that are near the global optimum; thousands or tens of thousands (or more) of operations may be required to identify solutions that are close to the global optimum, though, with the majority of conventional genetic algorithms (Lim et al
106). Based on its ability to be mass produced with relative ease, highly integrated field effect technology introduced a number of opportunities for configuring large numbers of field effect chips in parallel architectures (MacKenzie, 1998)
The speedup of code refers to the extent to which performance gain is achieved by using multiple processors to run code in parallel compared to a single processor (Sadjadi, 2009). Inspired by Charles Darwin's original concepts of biological evolution via the natural selection process that produces the "survival of the fittest," genetic algorithms seek to identify near-global optimal solutions from an initial set of guesses that are entirely random (Metcalfe & Charbonneau, 2003)
Benchmark Testing on Supercomputers Benchmark testing on supercomputers remains in its formative stages. At present, graph data has emerged as the gold standard for benchmark testing (Mims, 2010)
1). With respect to the type of architecture used for distributed computing applications, the primary objective is to employ various sizes of networks of processors in order to more readily solve large problems (Pan, Lubomir, Dillencourt & Lai, 2005)
It is also important to point out, though, that the "automatic" aspects of the design process are based on carefully formulated requirements that software engineers must develop based on the specific requirements of the initiative. In the case of the NASA space antenna, the "fitness criteria" for the design specified that the antenna must be able to transmit and receive signals broadcast within certain frequency parameters as well as the antenna's overall size; however, these fitness criteria to not instruct the algorithm as to what types of materials should be used or how they should be configured in the final design (Plotkin, 2009)
37). The speedup of code refers to the extent to which performance gain is achieved by using multiple processors to run code in parallel compared to a single processor (Sadjadi, 2009)