Here is another article asking whether multicore can scale or not, emphasizing the need for solving memory performance bottleneck. Jack Ganssle of Embedded.com, earlier this year, wrote an article titled, “Is multicore hype or reality?“. Here is another article by Jack.

“My February column in Embedded Systems Design Magazine was an attempt to show that the emperor, at least when talking about multicore technology, has no clothes. Multicore is being hyped as the solution to clock rate stagnation, when it really addresses two problems:

- A handful of “embarrassingly parallel” problems can derive great performance benefits from SMP.
- In many applications one can reduce power consumption by using more processors at slower clock rates.
Actually, there is a third problem that multicore solves: the vendors’ need to sell us more transistors as they continue to exploit Moore’s Law.

Now a study in IEEE Spectrum shows that even for the classic embarrassingly parallel problems like weather simulations multicore offers little benefit. The curve in that article is priceless. As the number of cores grow from two to 64 performance plummets by a factor of five. Additional processors nullify each other.

Call it the Nulticore Effect.”

Full Story

Related Posts
Is Multicore Bad News For Supercomputers?
Is more than 16 cores pointless?
Is Multicore The Way To Go?
Multi-Core a Drag on Some HPCs