Dual core technology refers to two individual microprocessors on a single die cast chip. This is essentially two computer processing units (CPUs) in one. The advantage of a dual core chip is that tasks can be carried out in parallel streams, decreasing processing time. This is referred to as thread-level parallelism (TLP).
TLP is also possible on motherboards that can accommodate two separate CPU dies. When TLP is accomplished in a single CPU through dual core technology, it is called chip-level multiprocessing (CLM).
In dual core CPUs, each microprocessor generally has its own on-board cache, known as Level 1 (L1) cache. L1 cache significantly improves system performance, because it is much faster to access on-chip cache than to use random access memory (RAM). L1 cache is accessed at microprocessor speeds.
Dual core chips also commonly feature secondary shared cache on the CPU, known as Level 2 (L2) cache. Motherboards may also have a cache chip designated as Level 3 (L3) cache. While faster than RAM, L3 cache is slower than cache built into the dual core chip.
For software to take advantage of dual core architecture, it must be written to utilize parallel threading. Otherwise, the program functions in single-core mode, using just one data stream or one of the built-in microprocessors. Unfortunately, coding for TLP is quite intensive, as interleaving shared data can create errors and slow performance. Because of these and other issues, a dual core processor does not deliver twice the speed of a single-core processor, though there is a significant increase in performance under optimal conditions. Finally, dual core chips run hotter than their single-core cousins.

0 comments:
Post a Comment