lecture 9.pptx

# lecture 9.pptx - Lecture 9 Parallel Computation Computation...

• Notes
• 29

This preview shows pages 1–8. Sign up to view the full content.

Lecture 9

This preview has intentionally blurred sections. Sign up to view the full version.

Parallel Computation Computation Structure for Parallel Computation op Start End 0 time 0 time Absorbing control node t time test Start control node Stop control node Operation node Test node Or node, also called Junction node Many in, One out For parallel computation, need to generate threads, one where a parent produces children: a fork node to create threads. The corresponding join node, also called “and” node or synchronization node , equivalently, is the node where parallel threads come together into one. It is called an and node because need all control signals to arrive at join before outgoing control signal generated.
These two nodes are designed to be put together with operations t n time Fork Join op1 opn . . . t 1 time In this parallel structure, each thread has its own time.

This preview has intentionally blurred sections. Sign up to view the full version.

Performance Analysis of Parallel Structures Fork C1 Cn Join e 1 e 1 = e 1 e 1 PS C=C FORK + C JOIN + max(C 1 , C 2 , …C n )
1 Fork Join 2 3 e 3 4 5 e 2 V Need to find the execution time of every branch (thread) Each branch is treated as a separate computation e 1

This preview has intentionally blurred sections. Sign up to view the full version.

Given the value of the flows e1, e2 What is the execution time of this branch? Total number of iterations is e2 The number of traversal of the branch is e1 The average number of iterations per traversal is e2/e1 The execution time of the branch = e2/e1 * C 5 + (1+e2/e1) * C 4 4 5 e 2 V e1
e1 Probability of the right branch = e3 /e1 Probability of the left branch = (e1-e3) / e1 The execution time of the IF statement = e3/e1 * C3 + (e1-e3)/e1 C2

This preview has intentionally blurred sections. Sign up to view the full version.

This is the end of the preview. Sign up to access the rest of the document.
• Spring '14
• Ammar,R

{[ snackBarMessage ]}

### What students are saying

• As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

Kiran Temple University Fox School of Business ‘17, Course Hero Intern

• I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

Dana University of Pennsylvania ‘17, Course Hero Intern

• The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

Jill Tulane University ‘16, Course Hero Intern