composite cores pushing heterogeneity into a core n.
Download
Skip this Video
Loading SlideShow in 5 Seconds..
Composite Cores: Pushing Heterogeneity into a Core PowerPoint Presentation
Download Presentation
Composite Cores: Pushing Heterogeneity into a Core

Loading in 2 Seconds...

play fullscreen
1 / 3

Composite Cores: Pushing Heterogeneity into a Core - PowerPoint PPT Presentation


  • 172 Views
  • Uploaded on

Composite Cores: Pushing Heterogeneity into a Core. Andrew Lukefahr , Shruti Padmanabha , Reetuparna Das, Faissal M. Sleiman , Ronald Dreslinski , Thomas F. Wenisch , and Scott Mahlke University of Michigan Micro 45 May 8 th 2012. Creating a Composite Core. iCache. Fetch. Decode.

loader
I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
capcha
Download Presentation

PowerPoint Slideshow about 'Composite Cores: Pushing Heterogeneity into a Core' - adeola


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.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
composite cores pushing heterogeneity into a core

Composite Cores:Pushing Heterogeneity into a Core

Andrew Lukefahr, ShrutiPadmanabha, Reetuparna Das, Faissal M. Sleiman, Ronald Dreslinski, Thomas F. Wenisch, and Scott Mahlke

University of Michigan

Micro 45

May 8th 2012

creating a composite core
Creating a Composite Core

iCache

Fetch

Decode

O3 Execute

iTLB

RAT

Load/Store Queue

dTLB

Reg File

Branch Pred

dCache

iCache

Fetch

dTLB

~50 KB

Controller

<1KB

~50KB

iTLB

dCache

Branch Pred

dCache

iCache

Fetch

Reg File

Decode

Mem

dTLB

iTLB

inO Execute

Branch Pred

little backend utilization
Little Backend Utilization

Traditional OS-Based Quantum

Fine-Grained Quantum

  • 3-Wide O3 (Big) vs. 2-Wide InOrder (Little)
  • 5% performance loss relative to all Big

More time on little backend with same

performance loss