A processor with asymmetrical cores is one in which the design of the cores is heterogeneous. Typically this means that, in relation to one another, each of the cores can be designed to operate with different instruction sets, clock speeds, and have differing memory and programming models. (Ke Ning, K; Yi, G; Gentile R., 2005) The key benefit of such a model is that each of the cores is typically specialized to accomplish a specific type of task; therefore, yielding improved performance; however, there are a number of disadvantages associated with this model as well. There are two main disadvantages for this model: first, relatively to processors with symmetrical cores, development of applications is very complex and more difficult; second, due to the specialization of the processors, one that is underutilized cannot be as easily leveraged to assist with general processing.
A processor with symmetrical cores is one in which the design of the cores is homogeneous. Unlike processors with asymmetrical cores, the cores contained in this type of processor are identical to one another and are intended to be used for all purposes and types of tasking. The advantage of these types of processors is that, because there is only one type of core design, developing applications for them is easier relative to processors with asymmetrical cores; Additionally, because the cores are generalized, unused processing power of one core can be more easily applied to accomplish the tasking of another. Naturally, the one obvious disadvantage of this model is that, because the cores are designed for general use, they cannot be optimized to perform a particular type of task.
Ne Ning, K; Yi, G; Gentile R. (2005). Single-Chip Dual-Core Embedded Programming. Models for Multimedia Applications. Ecnmag.com. February 2005. Retrieved March 5, 2008 fromhttp://www.ecnmag.com/single-chip-dual-core-embedded.aspx?menuid=580