1 / 10

Triggering Parallel Dimensions: Matrix Constructors and Propagation of Parallelism

This outline provides a comprehensive guide on implementing parallel dimensions in matrix constructors using specific front-end tags. It covers the configuration of matrices, commands for random generation, and the propagation of parallelism with a focus on handling confidential information. By utilizing parameters such as m and n, the outline details procedures for matrix creation and manipulation, ensuring efficient data handling in parallel computing contexts. The processing commands emphasize confidentiality while teaching how to manage and transmit data effectively.

redell
Télécharger la présentation

Triggering Parallel Dimensions: Matrix Constructors and Propagation of Parallelism

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. The “*p” Outline: Front end tag for Parallel Dimensions Trigger for matrix Constructors Propagator of Parallelism Confidential Information

  2. The “*p” Outline: Front end tag for Parallel Dimensions Trigger for matrix Constructors Propagator of Parallelism Confidential Information

  3. The “*p”: Front end tag for “parallel dimensions” Commands, pointers transmitted. Not the data! Confidential Information

  4. The “*p”: Front end tag for “parallel dimensions” >> m=100, n=200*p m = 100 n = 200p m,n Commands, pointers transmitted. Not the data! Confidential Information

  5. The “*p”: Front end tag for “parallel dimensions” >> m=100, n=200*p m = 100 n = 200p >> a=rand(m,n), b=rand(n,m) a = ddense object: 100-by-200p b = ddense object: 200p-by-100 m,n Build random m,n a,b return pointer Commands, pointers transmitted. Not the data! Confidential Information

  6. The “*p”: Front end tag for “parallel dimensions” >> m=100, n=200*p m = 100 n = 200p >> a=rand(m,n), b=rand(n,m) a = ddense object: 100-by-200p b = ddense object: 200p-by-100 >> np ans = 4 m,n Build random m,n a,b return pointer a: col dist 100x50 per processor b: row dist50x100 per processor Commands, pointers transmitted. Not the data! Confidential Information

  7. The “*p” Outline: Front end tag for Parallel Dimensions Trigger for matrix Constructors Propagator of Parallelism Confidential Information

  8. The “*p”: Trigger for Matrix Constructors >> m=100, n=200*p m = 100 n = 200p >> a=rand(100,200*p); a=randn(m,n); >> a=zeros(m,n); a=ones(m,n); >> a=eye(200*p); >> a=1:200*p; m,n Matrix Constructor m,n a return pointer Commands, pointers transmitted. Not the data! Confidential Information

  9. The “*p” Outline: Front end tag for Parallel Dimensions Trigger for matrix Constructors Propagator of Parallelism Confidential Information

  10. The “*p”: Propagator of Parallelism n >> n=2000*p; %Trigger constructors >> a=rand(3,n); >> x=rand(n,1); %Propagate Parallelism!!! >> k=size(x,1); >> b=ones(k); >> e=eye(size(a)); a x k b e k,n a,b,e,x Confidential Information

More Related