The Resurgence of Parallelism

Parallel computation is making a comeback after a quarter century of neglect. Past research can be put to quick use today.

Many SoC designers—and ARM licensees—are struggling on one issue in today’s environment of multiple GPUs and MPUs: Is there a compiler that can automatically spread load energy efficiently across those multiple cores in a SoC?

ARM’s answer is, no, not yet.

Acknowledging that “huge software challenges are coming,” Inglis said, “Many SoC vendors are, just about now, running into an ‘oops’ phase.” He warned that the industry is still in a very early stage of using multiple cores for their SoCs. “Real SoCs based on multiple cores to run real applications are just about to come out now. They are being debugged. Many engineers are suddenly asking, ‘how do you run all these tasks in so many cores?’”

I believe it is time for us to move from our deus ex machina model of explicitly managed resources to a fully distributed, asynchronous model that embraces component failure as a standard occurrence. To draw a biological analogy, we must reason about systemic, organism health and behavior rather than cellular signaling and death, and not allow cell death (component failure) to trigger organism death (system failure). Such a shift in world view has profound implications for how we structure the future of international high-performance computing research, academic-government-industrial collaborations and system procurements.

Perhaps the necessities of programming very (very) large systems may force some of the software issues currently hamstringing multicore.

The impact is akin to putting a Ferrari engine in a go-cart; the power may be there, but design mismatches severely limit the ability to exploit it.