Intel’s Chief technology officer Justin Rattner, who showed off a computer running two of the first four Clovertown chips produced, declined to say whether all four cores are on a single die, or if Clovertown would use two dual-core chips stuck together. Putting cores on the same die is more efficient since they can more easily exchange information and share resources.
Rattner said he envisioned a day when a single chip will have tens or even hundreds of cores, echoing the early era of the electronics industry when companies raced to see how many transistors they could squeeze on a chip. "It’s kind of like how many angels can you fit on the head of a pin?" Rattner said. But he said there were significant challenges in breaking beyond eight or 16 cores, from how to provide enough system memory to how to write software to take advantage of the new features.
Adding cores requires careful planning. Energy efficiency, data input/output and memory latency (the time it takes data to go from memory and the processor and vice versa) will be major issues with each level of core expansion.
To get around some of these issues, Intel is conducting research into circuit design and chip architecture as it has in the past. In addition, the company is working with application developers to determine how the architecture of its chips can be optimized.
By working with one server application developer, Intel determined that it needed to make three small changes to the architecture of one of its future server chips. Before the changes, the application only ran well in simulations on chips with 16 cores. After that, performance began to decline, Rattner said. After the changes, performance continued to climb. "We got it to scale well past 32" cores, he said.
