This blog aims at creating awareness of parallel computing crisis and coming up with ideas for the processor of 2020. By that time CPU design will see some fundamental changes. That change will probably be a breath of fresh air for the processor makers for the next 30 years or so.
Today we have a solution to the crisis: Multicores. But is this the only way to get the required computational power in scientific and consumer applications?
The fundamental question is how will the processor of 2020 look like, when we will reach the limit on core counts per chip, in addition to memory wall, ILP wall and the recent one being power wall?