Multicore going mainstream

Having attended the Mobile World Congress in Barcelona, and shortly after the Embedded World congress in Nuremberg, it's clear to me that multicore is happening, and it's happening fast. At the Mobile World Congress, nVidia showed off their Denver mobile processor with a quad core inside. Qualcomm announced their new Krait architecture, which includes up to four cores. Intel's Sandy Bridge started shipping in January, with four x86 cores on board. Haswell, Intel’s next architecture revision, defaults to eight cores. This weekend, a friend of mine showed me his new phone, the LG Optimus 2X. The 2X stands for dual core; multicore is even becoming a topic the marketeers touch.

Multicore is here since it solves several challenges:

Multicore solves the compute challenge

Many applications can be significantly sped up through parallelization. A media player for instance contains graphics, audio and video, each of which can be split up over multicore cores to give more performance. More graphics, audio and video performance means a better user experience.

Multicore solves the hardware design challenge

It's relatively easy to build multicore hardware. Replicate the design, then add an interconnect. It’s much simpler to implement a quad core processor, than to increase a single processor’s performance fourfold.

Multicore solves the power challenge

One way to make processors more powerful is to introduce more pipeline stages, and increase the clock rate. Driving up the clock requires more power. More pipeline stages means there’s less work done per stage. This isn’t a very power-efficient approach, and already some time ago frequencies have stopped scaling with new process technology nodes. Using multiple cores lowers the average clock frequency, reducing energy consumption.

Multicore is here, and here to stay. The crux is in the programming.