High Performance Computing: Tools and Applications
CSE 6530

Spring 2016
C530
Week 04 Highlight:
Optimal Input Usage Rules
Steven Kreft and Ellie MafiKreft
Thursday, April 7th, 9:00pm EST
Marginal Product of Labor
An important concept is the marginal product of
labor:
MPL = Q/L
This is the additional output the firm would
g
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
function []= clockFaces (time, dif)
%Pulling out the hour and the minutes and converting them to doubles
min= str2num(time(end1):end);
time(end2):end)=[];
hour=str2num(time);
min= min*6;
hour=hour*30;
sign= dif(1);
dif(1)=[];
dif=str2num(dif);
%Current
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
function [str1]= lostAbroad(arr, movea)
row= movea(1,:);%Seperating the row and column indication vectors
col= movea(2,:);
str1=[]; %Initializing str1
o= length(col);
pd=false;
ps=false;
for x=2:o %Starting at the second position in both row and column ve
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
function [str]= dragRace(t, v,arr,dist)
p= length(arr);
dis=[];%Initialization
timec=[];
accel=[];
for x=1:p
t1= t(x,:);
v1= v(x,:);
0umerical integrating to find the distance each car covers
distcov=cumtrapz(t1,v1);
cover= distcov(end);
dis=[dis, cover];
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
function [prime]= primeFinder (int)
a=round(sqrt(int);%Square root of integer
b=3;
p=[];
for x= 1:a; %Vector from 3 until the square root of the integer
if mod(int,x)=0 & x>1; %Checking divisibility of the integer by
interations of changing x
a=[false]; %
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
function []= plotSnake(arr)
%Setting the bounds of the board
[r,c]= size(arr);
x= [1,c,c,1,1];
y=[1,1,r,r,1];
plot(x, y, 'k'); %Sets the boundaries for the plot
hold on %Makes sure that all of the plots are on onem plot
axis([0 (c+1) (r1) 0]);
tit
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
OpenMP Examples OpenMP
Submitting Jobs and compiling
Accessing Hogwarts Accessing
You You
access the Hogwarts cluster through the head node Dumbledore. the [email protected]
ssh
Simple Example Code Simple
int main(int argc, char* argv[]) c
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
CS6230 HPC Tools and Applications Parallel NBody Applications
Jeffrey S. Vetter
Computational Science and Engineering College of Computing Georgia Institute of Technology
http:/ft.ornl.gov/~vetter [email protected]
With input from various ORNL sources
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
Introduction
Example problem
Code implementation
Learn more
An example to demonstrate how to use PETSc
(only very basic usage)
Wenjun Deng
January 2010
Introduction
Example problem
Code implementation
Learn more
Contents
1
Introduction Example problem Cod
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
% Test case generator for whenisGood
% Input:
1) Double containing the number of rows for the schedule array
(number of people)
%
% Outputs: 1) Mx24 array generated with availabilities and busy time
%
blocks. M is the input.
%
2) A char array containing i
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
function []= metaData (ca)
[num, txt,raw]=xlsread(ca);
[r,c]=size(raw);%Number of rows and columns
N= ceil(c1)/2); 0imensions of the sublot array
%LOOP FOR CONSTRUCTING EACH OF THE SUBPLOTS
for x= 2:c
xval= num(:,1);%Getting the x values
yval= num(:,x);%
High Performance Computing: Tools and Applications
CSE 6530

Spring 2016
C530
Week 03 Highlight:
Price Elasticity, the Total Revenue Test,
and Sales Targeting
Steven Kreft and Ellie MafiKreft
Thursday, March 31st, 9pm EST
Demand Elasticity
EQ , P
%QX d
%PX
Own Price elasticity describes how
responsive sales volume is to price
High Performance Computing: Tools and Applications
CSE 6530

Spring 2016
C530
Week 09 Highlight:
The Basics of Game Theory
Steven Kreft and Ellie MafiKreft
Thursday, May 12th 9:00pm EST
Basics of the Game
Players:
These are the decision makers within the game.
Usually within firms, government, or interest groups.
Strategi
High Performance Computing: Tools and Applications
CSE 6530

Spring 2016
C530
Week 08 Highlight:
Comparing Oligopoly models
Steven Kreft and Ellie MafiKreft
Thursday, May 5th, 9:00pm EST
Oligopoly Models
Your optimal price and output depends on:
Beliefs about the reactions of rivals.
Your choice variable (P or Q)
The natu
High Performance Computing: Tools and Applications
CSE 6530

Spring 2016
C530
Week 07 Highlight:
Managing in Different Market Structures
Steven Kreft and Ellie MafiKreft
Thursday, April 28th, 9:00pm EST
Perfect Competition
Characteristics:
Numerous buyers and sellers.
Homogeneous product.
Perfect Information.
No transaction
High Performance Computing: Tools and Applications
CSE 6530

Spring 2016
C530
Week 10 Highlight:
Market Power Pricing Strategies
Steven Kreft and Ellie MafiKreft
Thursday May 19th, 9:00pm EST
Market Power Pricing
Market Power permits firms to sustain prices
above marginal costs.
How much the firm can mark up price over cost
High Performance Computing: Tools and Applications
CSE 6530

Spring 2016
C530
Week 02 Highlights
Supply, Demand, and Market
Adjustments
Steven Kreft and Ellie MafiKreft
Thursday, March 24th, 9:00pm EST
Demand and Supply
Demand: Willingness to buy function that is
based on the Value to Consumers.
Consumer Surplus (CS): the d
High Performance Computing: Tools and Applications
CSE 6530

Spring 2016
C530
Week 05 and 06
Highlights
Steven Kreft and Ellie MafiKreft
Thursday, April 14th, 9:00pm EST
Week 05 Highlight:
Market Structure, Conduct, and
Performance
Industry Analysis
Industry Structure
Number of firms, Size of firms, Leader/Follower
aspects,
High Performance Computing: Tools and Applications
CSE 6530

Spring 2016
FlexIO: Locationflexible Execution of In Situ
Data Analytics for Large Scale Scientific
Applications
Fang Zheng, Hongbo Zou, Greg Eisenhauer, Karsten Schwan, Matthew Wolf, Jai Dayal,
TuanAnh Nguyen, Jianting Cao, Hasan Abbasi*, Scott Klasky*, Norbert Po
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
CS6230 HPC Tools and Applications Scientific Frameworks PetSC
Jeffrey S. Vetter
Computational Science and Engineering College of Computing Georgia Institute of Technology
http:/ft.ornl.gov/~vetter [email protected]
Motivation
First few months of this co
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
Announcements
Should be working on your projects
Midproject status reports will be requested during week of Nov 1
Schedule
26Oct10
[email protected]
1
CS6230 HPCToolsandApplications ReductionsandParallelPrefixSum
JeffreyS.Vetter
ComputationalScience
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
Announcements
Registration is full and almost closed Class schedule
Class on Aug 30, Sept 1 NO class on Sept 3, 6 (Holiday), 8 Class on Sept 10
TA
Jonathan Brownsworth [email protected]
Action Items
Brainstorming term project ideas and collab
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
A decadal DOE plan for providing exascale applications and technologies for DOE mission needs
Rick Stevens and Andy White, cochairs
Pete Beckman, Ray BairANL; Jim Hack, Jeff Nichols, Al GeistORNL; Horst Simon, Kathy Yelick, John ShalfLBNL; Steve Ashby,
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
CS6230: HPC Tools and Applications
Term Project Discussion Jeffrey Vetter
Term Project (1)
The term project will focus on a research or application related HPC project that will depend greatly on the students skills and status.
Students will identify te
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
Announcements
Class schedule
Class on Aug 25, 27, 30, Sept 1 NO class on Sept 3, 6 (Holiday), 8 Class on Sept 10
TA
Jonathan Brownsworth [email protected]
Homework #1
Survey at http:/www.surveymonkey.com/s/ SRTJ7GL Submit the survey at this w
High Performance Computing: Tools and Applications
CSE 6530

Fall 2010
CS(E)6230: HPC Tools and Applications
#01: Introduction Jeffrey Vetter
Outline
My background Course Overview A brief history of supercomputing Toward Exascale computing
Computational Science and Engineering Division
2
My Background
Joint Professor w/ Oa