Imperial College London > Talks@ee.imperial > CAS Talks > Increasing the efficiency and accessibility of next-generation computing platforms

Increasing the efficiency and accessibility of next-generation computing platforms

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Grigorios Mingas.

Next generation processors increasingly leverage parallelism, instead of significantly increasing the clock speed, to provide greater computing power. Thus, to achieve greater processing efficiency, the objective becomes the effective mapping of tasks onto modern and future multicore computing platforms. Challenges that complicate this goal include the dynamic and unpredictable nature of workload interactions and the dependency of an application’s behaviour on its input data. For asymmetric and heterogeneous multicore architectures used in embedded systems and (future) data centers, the difficulty of effectively mapping computing tasks to achieve high processing throughput while ensuring quality of service is further magnified.

This talk will discuss our research efforts using FPG As to:

1) create a scalable multicore framework with OS support for assymmetric, heterogeneous, and reconfigurable computing architectures;

2) enable programmers to understand their program’s execution on a computing platform at runtime;

3) allow the scheduler to dynamically recognize workload requirements and tailor its task assignments to processing cores accordingly; and

4) abstract heterogeneous platforms with hardware accelerators from software developers via the OS scheduler.

This talk is part of the CAS Talks series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

Changes to Talks@imperial | Privacy and Publicity