Multicore scaling is a computer architecture approach that increases overall processing performance by integrating a greater number of independent processor cores onto a single chip rather than increasing the clock frequency or complexity of an individual core. This design strategy arose as physical power and thermal limits, particularly the end of Dennard scaling, restricted the ability to continuously speed up single-threaded processors. By dividing parallel workloads across multiple cores, multicore scaling seeks to sustain generational computing gains alongside increasing transistor density, though its practical limits are governed by power consumption constraints, heat dissipation, memory communication bottlenecks, and the degree of inherent parallelism in software.
