1 / 11

Energy-efficient Microkernel

Energy-efficient Microkernel. 06/06. Progress. Paper Studied Energy-Efficient Mapping Technique for Virtual Cores Convex Optimization in Local Single-Threaded Parallel Mobile Computing Real-Time Dynamic Voltage Loop Scheduling for Multi-Core Embedded Systems. Scenario.

emelda
Download Presentation

Energy-efficient Microkernel

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Energy-efficient Microkernel 06/06

  2. Progress • Paper Studied • Energy-Efficient Mapping Technique for Virtual Cores • Convex Optimization in Local Single-Threaded Parallel Mobile Computing • Real-Time Dynamic Voltage Loop Scheduling for Multi-Core Embedded Systems

  3. Scenario • Each processor has a memory queue that accommodates an arbitrary maximum number of tasks. • Tasks and processors are heterogeneous.

  4. Cost Function • Cj: cost of the jth stream • ij : # of task in queue • ε%,j : remaining power • Al,j : availability of executing Tl in the jthstream • tθ,r,j : overhead access time of a task Tr to be accessed by Pj

  5. Minimizing Constrained Cost Function

  6. Simplification • Memory queue: no limitation • Single power source • Remove stochastic availability • Discrete processing rate

  7. Some Math • εk=Bk(pk)2, pk = pH or pL • εk,H = 4 * εk,L • Change pk from pL → pH if • αε,k(4 * εk,L- εk,L) < • “Is it always true that we should adjust Pm before Pn, m < n?”

  8. Cutting Point • Assume that we can change the order of tasks in queue. • Can we find a cutting point? • Cutting point: there will be not benefit if we switch the processing rate from pLto pH after that task. pH pL

  9. Two Processing Rates • Initially all N tasks use processing rate pL. • pH = 2* pL • Sorted according to task execution time. • Descending order. • Started from task 1, change the processing rate of task k from pLto pH and compute the benefit.

  10. Two Processing Rates(Cont.)

  11. Next • Multiple processing rates • Consider that each task has different cost factors.

More Related