1 / 18

Computer Architecture SIMD

Computer Architecture SIMD. Ola Flygt Växjö University http://w3.msi.vxu.se/users/ofl/ Ola.Flygt@msi.vxu.se +46 470 70 86 49. Outline. Design space Granularity Connectivity Processor complexity Local Anotomy Fine grained architectures Course grained architectures. CH01.

abena
Download Presentation

Computer Architecture SIMD

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. Computer ArchitectureSIMD Ola Flygt Växjö University http://w3.msi.vxu.se/users/ofl/ Ola.Flygt@msi.vxu.se +46 470 70 86 49

  2. Outline • Design space • Granularity • Connectivity • Processor complexity • Local Anotomy • Fine grained architectures • Course grained architectures CH01

  3. A typical SIMD system

  4. The development of SIMD systems

  5. Design space • There are several factors to consider when designing a SIMD. The most important ones are: • Granularity • Connectivity • Complexity of processing element • Local autonomy

  6. Granularitymapping data to PEs

  7. GranularityFine to coarse grain

  8. Connectivity Not all the types of connectivity for data parallel arch. are used in SIMD arch.

  9. Performance of interconnection networks • There are several important parameters in an interconnection network • Diameter, the shortest path between the two most distant nodes • Bandwidth, how many data exchanges simultaneous in the network • Blocking, how many exchanges are blocked

  10. Performance of interconnection networks N=128*128 1 1 256 28 14 14 1 N=128*128 1 16384 16384 28 14 16384 16384

  11. Processor complexity • The design space for processor precision • Single bit – fine grained, mainly for image processing • Integer – general computing • Floating point – Coarse grain, scientific computing • Fine grained -> more PEs

  12. Local autonomy

  13. Fine grained SIMDThe MPP system

  14. Fine grained SIMDThe MPP processing element

  15. Fine grained SIMDThe MPP integrated circuit

  16. Coarse grained SIMDThe CM5 processing element

  17. Coarse grained SIMDThe CM5 fat tree connectivity

  18. Coarse grained SIMDThe CM5 system parameters

More Related