Meeting the challenges in computer architecture

Professor Per Stenström was one of eight Swedish scientists who recently was rewarded with an ERC Advanced Grant from the European Research Council – a grant whose primary purpose is to support and encourage the very best, truly creative scientists to be a bit more adventurous and take risks in their research.
Per Stenström's new project Mecca, financed by the ERC Advanced Grant, takes a novel, interdisciplinary and unconventional approach to address three important challenges facing computer architecture: parallelism, power and predictability.

Unlike earlier, predominantly disciplinary approaches, Mecca bridges layers in computing systems from the programming language/model, to the compiler, to the run-time/OS, down to the architecture layer. This opens up for exchanging information across layers and use it to manage parallelism and architectural resources automatically “under the hood” to meet challenging performance, power, and predictability requirements for future computers systems.

Software productivity threatened
Computer technology has doubled computational performance biannually over the past several decades. This performance growth rate has been an enabler for the dramatic innovation in information technology that now embraces our society. Before 2004, application developers could exploit this performance growth rate with no effort. However, since 2004 power consumption of computer chips exceeded the allowable limits and from that point and onwards, parallel computer architectures became the norm. Currently, parallelism is completely exposed to application developers and managing it is difficult and time-consuming. This seriously cuts down software productivity, which can stall progress in information technology.

The challenges
Technology forecasts predict that by 2020 there will be hundreds of processors on a computer chip. Apart from the challenge to manage parallelism on such architectures, keeping power consumption within allowable limits will remain a key roadblock for maintaining historical performance growth rates. Power efficiency must increase by at least an order of magnitude in the next ten years to not limit the growth rate. Finally, as computer chips are universal components, they are not only used as number-crunching devices in supercomputer and data centers but are increasingly being used as key components in embedded controllers in e.g. automotive applications where stringent timing responses are mandatory. Delivering predictable and tight response times using parallel architectures is a challenging and unsolved problem.

Project information
The Mecca project, Meeting the challenges in computer architecture, is financed by ERC through an Advanced Grant. The official starting date is February 1st 2014, and the grant is for 2,3 million Euro over 3 years.

Read the Press release from ERC

Read information on ERC Advanced Grants

Contact
Per Stenström, Department of Computer Science and Engineering, +46 31-772 17 61, per.stenstrom@chalmers.se
 
Text: Anneli Andersson
Photo: Peter Widing

Published: Thu 10 Apr 2014.