Chip Multicore Processor (CMP) has become the mainstream microprocessor architecture in nowadays industry and academic literature. In CMP processor, multiple traditional superscalar or simple cores are integrated into one silicon die. Such organization can exploit both instruction level parallelism (ILP) and thread level parallelism (TLP) in a program. Today, dual core microprocessors can be seen in everywhere, from supercomputer to server, desktop, laptop, and many embedded devices. Recently, Intel has announced their quad core microprocessor, Xeon 5300 [1] for high end server and workstation market and QX6700 [2] for low end desktop market. Also, a further development of multicore processor is the many-core processor, for example, the 80cores prototype in Intel [3] and GPGPU [4] in NVidia which has up to more than two hundreds stream cores on the chip. In academic research, interests have also moved to multicore and many-core processors. In computer architecture related top international conferences, ISCA, HPCA, Micro, ASPLOS, PACT, more than 80% of the papers are direct or indirect related to multicore or many-core architectures. It is obvious that we have come to multicore and many-core era.