Register now to access 7 million high quality study materials (What's Course Hero?) Course Hero is the premier provider of high quality online educational resources. With millions of study documents, online tutors, digital flashcards and free courseware, Course Hero is helping students learn more efficiently and effectively. Whether you're interested in exploring new subjects or mastering key topics for your next exam, Course Hero has the tools you need to achieve your goals.
2 Pages
homework5
Course: CS 6290, Fall 2008School: Georgia Tech
Rating:
Word Count: 425
Document Preview
INSTITUTE GEORGIA OF TECHNOLOGY College of Computing CS6290/CS4290 High-Performance Computer Architecture Spring 2003 CS6290/CS4290 Homework 5 Issued: March 8, 2003 Due: April 4, 2003 Purpose: Reading: This homework introduces issues in network performance. H&P Chapter 8 [Kubi93] paper 1. Eective Bandwidth. 2. Topology. 3. Reading: [Kubi93]. (As in the syllabus) collaboration on projects and homework...
Unformatted Document Excerpt
Coursehero >> Georgia >> Georgia Tech >> CS 6290Course Hero has millions of student submitted documents similar to the one
below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.
Course Hero has millions of student submitted documents similar to the one
below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.
INSTITUTE GEORGIA OF TECHNOLOGY
College of Computing CS6290/CS4290 High-Performance Computer Architecture Spring 2003
CS6290/CS4290 Homework 5
Issued: March 8, 2003 Due: April 4, 2003
Purpose: Reading:
This homework introduces issues in network performance. H&P Chapter 8 [Kubi93] paper 1. Eective Bandwidth. 2. Topology. 3. Reading: [Kubi93]. (As in the syllabus) collaboration on projects and homework in pairs is encouraged. If you work in a pair, turn in one write-up with the names of both collaborators. Youre welcome to discuss high-level concepts with other groups, but all homework solutions must be worked out and written up separately.
Problems:
Collaboration:
Problem 1: Eective Bandwidth A: B: C: D: Problem 8.1 in the book. Problem 8.2 in the book. Problem 8.3 in the book. Problem 8.4 in the book.
CS6290/CS4290 Spring 2003
1
March 8, 2003
Problem 2: Topology Assume you have to connect 64 nodes, each with and input and an output link. You have available 16x16 switches that each have 16 inputs and 16 outputs. Assume that the latency through a switch (or node) is one unit of time and that the bandwith of a link is one unit. A: The lowest-cost network has no switches instead you connect the 64 nodes into a ring. What is the bisection bandwidth of the network? What is the average message latency between randomly selected pairs nodes? of B: Using 16x16 switches, design a switch topology that has the minimum latency through the switches. Sketch the topology. How many switches are required? What is minimum latency? What is the bisection bandwidth of your topology?
Problem 3: Reading Read the [Kubi93] paper on the Alewife message passing mechanism. A: Figure 8 shows the instruction sequence for composing and launching a message with a one-word payload of data where the data is written directly out of a register. Show the sequence for sending a message with an additional payload of 256 bytes from location foo in memory. B: Figure 8 also shows the detailed pipeline timing for launching a message: three cycles per stio plus one cycle for the ipilaunch plus two cycles of overhead. The processor runs at 33MHz; the network is byte-wide an...
Find millions of documents on Course Hero - Study Guides, Lecture Notes, Reference Materials, Practice Exams and more.
Course Hero has millions of course specific materials providing students with the best way to expand
their education.
Below is a small sample set of documents:
Below is a small sample set of documents:
Georgia Tech - CS - 2003
Articial Intelligence More Knowledge Representation and Reasoning (Approaches Other Than Logic)Nilsson - Chapters 18 Russell and Norvig - 10.5, 10.6Knowledge Representation and Reasoning; page 1 of 18planning operatorsFORALL s, x, store: At(st
Georgia Tech - CS - 4600
Articial Intelligence More Knowledge Representation and Reasoning (Approaches Other Than Logic)Nilsson - Chapters 18 Russell and Norvig - 10.5, 10.6Knowledge Representation and Reasoning; page 1 of 18planning operatorsFORALL s, x, store: At(st
Georgia Tech - CS - 2003
Classical Problem SolvingKnowledge Systems EngineeringCS 7613 Spring 2002Administrivia Class photo Mailing list? Laptop policyToday Skills survey results Getting and loading the BPS code Homework 0 Classical Problem Solvers2.80 2.60 2.
Georgia Tech - CS - 7613
Classical Problem SolvingKnowledge Systems EngineeringCS 7613 Spring 2002Administrivia Class photo Mailing list? Laptop policyToday Skills survey results Getting and loading the BPS code Homework 0 Classical Problem Solvers2.80 2.60 2.
Georgia Tech - CS - 2002
CS 4210 - Advanced Operating Systems - Summer 2002Final ExamThere are 110 points on this exam. Your grade will be measured as a percentage over 100 points, so in effect there are up to 10 extra credit points available. Point values are in () for ea
Georgia Tech - CS - 4210
CS 4210 - Advanced Operating Systems - Summer 2002Final ExamThere are 110 points on this exam. Your grade will be measured as a percentage over 100 points, so in effect there are up to 10 extra credit points available. Point values are in () for ea
Georgia Tech - CS - 4803
Problem 1(a) The forward and backward errors when using sin(x) x are shown below: x forward backward 0.1 0.000167 0.000167 0.5 0.0206 0.0236 1.0 0.1585 0.571 (b) The forward and backward errors when using sin(x) x x3 /6 are shown below: x forward
Georgia Tech - CSE - 6740
CSE 6740: Problem Set 1Problem Set 1: Density Estimation, Regression, and ClassicationDue: 5PM Friday, Oct. 10, 2008In this problem set, you will use FASTlib, a C+ library for machine learning algorithm development, to implement several algorith
Georgia Tech - CSE - 6740
CSE 6740 Lecture 17How Can I Reduce/Relate the Data Points? (Association and Clustering)Alexander Gray agray@cc.gatech.eduGeorgia Institute of Technology1 / 38Today1 2Associations ClusteringCentral tasks for data mining.2 / 38Associ
Georgia Tech - CSE - 6740
CSE 6740 Lecture 12What Loss Function Should I Use? (Estimation Theory)Alexander Grayagray@cc.gatech.eduGeorgia Institute of TechnologyCSE 6740 Lecture 12 p.1/33Today1. Robustness ("How safe/stable is my loss function?") 2. Comparing Estim
Georgia Tech - CSE - 6740
CSE 6740 Lecture 25How Do I Evaluate High-Dimensional Integrals? (Sampling)Alexander Gray agray@cc.gatech.eduGeorgia Institute of Technology1 / 32Today1 2 3Integration and Sampling Monte Carlo Variance Reduction Markov Chain Monte Carlo
Georgia Tech - CSE - 6740
CSE 6740 Lecture 10What Loss Function Should I Use? (Maximum Likelihood and Bayesian Inference)Alexander Grayagray@cc.gatech.eduGeorgia Institute of TechnologyCSE 6740 Lecture 10 p.1/37Today1. Bayesian estimation 2. Should I be a Bayesian
Georgia Tech - CSE - 6740
CSE 6740 Lecture 23How Do I Optimize Convex/Linear Functions? (Unconstrained Linear Optimization)Alexander Gray agray@cc.gatech.eduGeorgia Institute of Technology1 / 30Today1 2 3Convex Optimization Problems Unconstrained Optimization: Lin
Georgia Tech - CSE - 6740
CSE 6740: Problem Set 2Problem Set 2: PCA, Kernels, and Graphical ModelsDue: 5PM Monday, Nov. 10, 2008First, download the datasets. All of the implementations should be done in FASTlib and submitted. Please be clear about which implementations c
Georgia Tech - CSE - 6740
CSE 6740 Lecture 21How Do I Learn Actions From Data? (Reinforcement Learning)Alexander Gray agray@cc.gatech.eduGeorgia Institute of Technology1/1Today1 2 3Reinforcement Learning RL Methods: Known Environment RL Methods: Unknown Environmen
Georgia Tech - CSE - 6740
FASTlib: A library for Fundamental Algorithmic and Statistical ToolsRyan Riegel and Nishant Mehtarriegel@cc.gatech.edu, niche@cc.gatech.eduFASTlab, Georgia Institute of TechnologyFASTlib: A library for Fundamental Algorithmic and Statistical To
Georgia Tech - CSE - 6740
CSE 6740 Lecture 14How Can I Learn Fancier (Nonlinear) Models? (Kernelization)Alexander Gray agray@cc.gatech.eduGeorgia Institute of Technology1 / 42Today1 2How to make more complex models using kernels Theory motivating kernels2 / 42
Georgia Tech - CSE - 6740
CSE 6740 Lecture 22How Do I Optimize the Parameters? (Unconstrained Optimization)Alexander Gray agray@cc.gatech.eduGeorgia Institute of Technology1 / 34Today1 2Unconstrained Optimization: General Unconstrained Optimization: Sum-of-Squares
Georgia Tech - CSE - 6740
CSE 6740 Lecture 18How Can I Reduce/Relate the Features? (Dimension Reduction)Alexander Gray agray@cc.gatech.eduGeorgia Institute of Technology1 / 66Today1 2Dimensionality Reduction: Linear Dimensionality Reduction: Nonlinear2 / 66Dim
Georgia Tech - CSE - 6740
CSE 6740 Lecture 16The Beautiful World of Graphical ModelsGuest Lecturer: Nishant MehtaGeorgia Institute of TechnologyWhat are graphical models?For a set of random variables, a graphical model eciently exhibits the dependence relationships of
Georgia Tech - CSE - 6740
CSE 6740 Lecture 26How Do I Evaluate Deeply-Nested Sums? (Graphical Model Inference)Alexander Gray agray@cc.gatech.eduGeorgia Institute of Technology1 / 48Today1 2 3Graphical Model Computations Exact Inference Algorithms Approximate Infer
Georgia Tech - CSE - 6740
CSE 6740 Lecture 20Hidden Markov Models: Modeling, Inference, LearningGuest Lecturer: Nishant MehtaGeorgia Institute of TechnologyWhat Are Hidden Markov Models?Hidden Markov models (HMMs) are discrete Markov processes where every state generat
Georgia Tech - CSE - 6740
CSE 6740 Lecture 27How Do I Evaluate Large Sums and Searches? (Generalized N-Body Problems)Alexander Gray agray@cc.gatech.eduGeorgia Institute of Technology1 / 69Generalized N-Body ProblemsGeneralized N-Body problems (GNPs), are typically co
Georgia Tech - CSE - 6740
CSE 6740 Lecture 24How Do I Optimize With Constraints? (Constrained Optimization)Alexander Gray agray@cc.gatech.eduGeorgia Institute of Technology1 / 46Today1 2Unconstrained Optimization: Latent-Variable Constrained Optimization2 / 46
Georgia Tech - CSE - 6740
Neural Networks, Another brick in the wall Guest Speaker: Nick the Greek OutlineGood Books on Neural Networks Todays LectureNeural Networks, Another brick in the wallPink FloydHow do neural networks look?Neuron (Perceptron), The fundamental El
Georgia Tech - CSE - 6740
FASTlib Design and Development ManualVersion 0.11Fundamental Algorithmic and Statistical Tools Laboratory (FASTlab) College of Computing Georgia Institute of Technology Atlanta, GA Director: Alexander Gray Core designers: Garry Boyer, Ryan Riegel,
Georgia Tech - CS - 4331
=Introduction to Problem Solving and Learning=Intelligent Agents:An Intelligent Agent is situated in an enviornment and interacts withit - i.e., it perceives the state of the environment and changes in,and acts on the environment to effect
Georgia Tech - CS - 3240
CS 3240: Languages and ComputationTuring Machine Acceptance ProblemFundamental QuestionsAre all problems Turing recognizable? Are all problems Turing decidable? The answers are no. But how do we prove it?Comparing Sizes of Infinite Sets
Georgia Tech - CHEM - 6472
CHEM 6491: Quantum Mechanics Problem Set IIDue Thursday, September 12 1. Find the eigenvalues and eigenvectors of 3 4 2 A = 0 1 2 0 0 0 and check that 1 + 2 + 3 = Tr(A) and 1 2 3 = det(A). 2. Which of the following are Hermitian? unitary? (a) (b
Georgia Tech - CHEM - 6472
CHEM 6472: Quantum Chemistry and Molecular Spectroscopy Problem Set IIIDue Tuesday, September 17 1. For an arbitrary normalized state | and operator A, calculate the sum the complete basis {|ki }. What value is obtained if A is unitary? 2. Show by
Georgia Tech - CHEM - 1310
CHAPTER 5: GasesChemistry of Gases Pressure and Boyles Law Temperature and Charles Law The Ideal Gas Law Chemical Calculations of Gases Mixtures of Gases Kinetic Theory of Gases Real GasesCHEM 1310 A/B Fall 2006Gases The states of matter: Gas:
Georgia Tech - CHEM - 1310
CHAPTER 12: Redox Reactions and Electrochemistry Recall GERtrude and LEO Gain of Electrons Reduction Loss of Electrons Oxidation Goals of Chapter: Understand redox reactions in detail Review oxidation numbers Learn electrochemical techniques
Georgia Tech - CHEM - 6472
CHEM 6472: Quantum Mechanics IR Spectroscopy Sample Questions(Not to turn in) 1. For a nonlinear molecule with N atoms, how many vibrational modes does it have? 2. Explain why a molecule with low symmetry is likely to have all its IR modes active. 3
Georgia Tech - CHEM - 1310
CHAPTER 13: Electrochemistry and Cell Voltage In this chapter: More about redox reactions Cells, standard states, voltages, half-cell potentials Relationship between G and voltage and electrical work Equilibrium constants from electrochemistry
Georgia Tech - CHEM - 1310
CHAPTER 11: Spontaneous Change and Equilibrium Goal of chapter: Be able to predict which direction a reaction will go (cases where there is not necessarily an equilibrium) At high temperatures, ice always melts instead of getting colder. (This is en
Georgia Tech - PHYSICS - 2021
THE SOLAR SYSTEM Physics 2021 Fall 2008Time: Instructor: Office: Email: Web Page: Text: 11 MWF J. R. Sowell W102 Physics Building jim.sowell@physics.gatech.edu http:/www.astronomy.gatech.edu Pathways to Astronomy, by Schneider & Arny COURSE POLICY
Georgia Tech - PHYSICS - 2021
PROBLEM SET 2 SOLUTIONSPhysics 20211. If phases are caused by the Earth's shadow, the curved terminator on the crescent and gibbous moons would indicate that both edges of the Earth are curved. However, at the quarter phases the terminator is strai
Georgia Tech - PHYSICS - 2021
PROBLEM SET 3 SOLUTIONSPhysics 20211. 1.4 deg x 60 arcmin/deg = 84 arcmin 84 arcmin x 60 arcsec/arcmin = 5040 arcsec d = 206,265 (D / ) a. b. c. d = 206,265 [( 2.6 cm ) / ( 3600 arcsec/deg )] = 150 cm = 1.50 m d = 206,265 [( 2.6 cm ) / ( 60 arcsec/
Georgia Tech - PHYSICS - 2021
PROBLEM SET 5 SOLUTIONSPhysics 20211. d = c t = (3 x 105 km/s) (4 hr) (3600 sec/hr) = 4.3 x 109 km c = f Fs 1 / ds2 = c / f = (3 x 108 m/s) / (3 x 1016 /s) = 1 x 10-8 m = 10 nm (X-ray) Fe 1 / de22.3.Fs / Fe = de2 / ds2 = (1 / 10)2 = 0.01 t
Georgia Tech - PHYSICS - 2021
PROBLEM SET 1 SOLUTIONSPhysics 20211. = 14 x 109 years x 365.25 days/year x 24 hr/day x 60 min/hr x 60 sec/min = 4.42 x 1017 seconds2a. 2b.Yes, if you are standing at one of the poles, all of the stars are circumpolar. Yes, if you are standing
Georgia Tech - PHYSICS - 2021
PROBLEM SET 11 SOLUTIONSPhysics 20211. The time between oppositions is the synodic period. E = 1 yr S-1 = E-1 P-1 S-1 = 1 (1.88) -1 = 0.468 S = 2.14 years (Earth Orbits) P = 1.88 yr P-1 = E-1 S-12.14 / 1.88 = 1.14 (Mars Orbits)2.Mars appea
Georgia Tech - PHYSICS - 2021
PROBLEM SET 7Physics 20211. A blackbody is brightest at a wavelength of 800 nm. At what wavelength would the blackbody be brightest if its temperature were tripled? 2. A blackbody has a temperature of 5 K. At what wavelength is it brightest? 3. Two
Georgia Tech - PHYSICS - 2021
PROBLEM SET 10Physics 20211. Find the largest angular size that Mercury can have as seen from the Earth. In order for Mercury to have this apparent size, at what point in its orbit must it be? 2. Suppose you have a superb telescope that can resolve
Georgia Tech - PHYSICS - 2021
PROBLEM SET 4Physics 20211. What is the orbital semi-major axis of a planet with a sidereal period of 25 years? 2. An asteroid is 2 AU from the Sun at perihelion and 6 AU from the Sun at aphelion. (a) Find the semi-major axis of the asteroids orbit
Georgia Tech - PHYSICS - 2021
PROBLEM SET 9Physics 20211. The angular diameter and distance of the Moon are 31 arcminutes and 384,400 km. Use the small angle equation to find the linear diameter of the Moon. 2. What is the angular diameter of the Earth as seen from the Moon? 3.
Georgia Tech - PHYSICS - 2021
PROBLEM SET 13Physics 20211. Demonstrate that Callisto obeys Kepler's Third Law. 2. Suppose the orbital period of one moon is exactly five-fourths as long as the orbital period of another. How often would the two moons pass one another as they orbi
Georgia Tech - PHYSICS - 2021
PROBLEM SET 12Physics 20211. What percentage of all planetary material in the Solar System is contained in Jupiter? 2. What is the ratio of Jupiter's volume to the volume of the Earth? 3. Using orbital data for a Galilean moon of your choice, calcu
Georgia Tech - MATH - 4032
MATH 4032 Test 2 Solutions, April 11, 2008, WTT Problem 1. Let n 3 and let Sn denote the shift graph with index set {1, 2, . . . , n}. Show that if k is a positive intger and n 2k , then (Sn ) k. Proof. We show that the shift graph Sn has a k-colo
Georgia Tech - MATH - 3012
MATH 3012 Homework Problems and Solutions Chapter 2, Spring 2009, WTT and MTKNote: These problems are grouped into two categories. The rst set consists of basic problems that everyone who expects to pass the course should be able to do. The second s
Georgia Tech - MATH - 2605
Solution for selected problems from chapter 4 Section 1: Problem 1 3 6 Problem 3 1 1 1 = 2 1 1 2 0 7 1 1 1 1 1 1 2 3 4 = 7 13 3 2 3 0 1 2 10 1 13 3 3 2Section 2: Problem 1 |z| = 4 2Problem 3 z= z1 z2 , z = z2 z1|z|2 = |z1 |2 + |z2 |2 = |z
Georgia Tech - MATH - 2605
Pratice Test 1B for Math2605, Spring 2004 Problem 1 a) Calculate the gradient of the function f (x, y, z) = x4 + y 4 + z 4 + xyz at the point (1, 2, 1). b) Find the linear approximation of the function f (x, yz) at the point (1, 2, 1). c) Calculate t
Georgia Tech - MATH - 2605
Pratice Test 1A for Math2605, Spring 2004 Problem 1 Consider the function f (x, y) = x4 + y 4 4xy . a) Calculate the gradient at the point x0 = (21/4 , 0). b) Find the equation for the line tangent to the curve f (x, y) = f (x0 ). c) Find all the po
Georgia Tech - MATH - 2605
Pratice Test 1A for Math2605, Spring 2004 Problem 1 Find the maximum value of the function 1+ x2 1 + (y 1)2on the set given by all pairs (x, y), such that x2 y 2 1. Find all the points where the maximal value is attained. Problem 2 a) Apply one
Georgia Tech - MATH - 2605
Solutions to practice test 3B Problem I: a) 1 1 2 Q= 2 1 5 1 5 7 R= 5 0 9 b) 1 5 1 + 3i U= 35 1 3i 5 U AU = c) eAt = 1 sin(3t) + 3 cos(3t) 5 sin(3t) 3 2 sin(3t) sin(2t) + 3 cos(3t) 3i 119 + 42i 0 105iProblem II: a) x (t) = b) T(t) = sin t co
Georgia Tech - MATH - 2605
Chapter 4 of Calculus+ : The Nonsymmetric Eigenvalue Problemby Eric A Carlen Professor of Mathematics Georgia Techc 2003 by the author, all rights reserved 1-1Table of Contents Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Georgia Tech - MATH - 4317
Analysis HomeworkSoniya Khan, Russel Krenek, Brendy Gallien, Matias Perez November 13, 20081) 1 proof: 2 0 1 2 Let the partition P = n , n , ., n1 , 1 . Therefore we have that xi = n 1 xi xi1 = n Now since f (x) is an increasing function on the in
Georgia Tech - MATH - 4280
MATH 4280 Hurley homework problem 2008February 27,1. There is test, commonly referred to as the quadruple test, given to pregnant women during the second trimester to detect fetuses with an increased risk of Downs Syndrome, among other conditi
Georgia Tech - MATH - 4280
MATH 4280 Hurley homework problem January 30, 20081 1. Let S be a source with probabilities 1 , 1 , 1 , 12 . 3 3 4(a) What is the average word length of a binary optimal code for S? (b) What are the possible word lengths l1 , l2 , l3 , l4 for a
Georgia Tech - MATH - 4280
MATH 4280 Hurley homework problem February 1, 20081. Find an r-ary Human encoding for a source with the probabilities pi = 0.3, 0.2, 0.2, 0.1, 0.1, 0.1 for r = 3 and r = 4 2. Find an r-ary Human encoding for a source with the probabilities pi = 0
Georgia Tech - MATH - 3012
MATH3012 Review for Test 1 Test 1 will cover the following sections: 1.11.4, 4.14.5, 5.15.3 and 5.5. The test problems will be similar to the problems assigned and examples given in the class. A formula sheet is allowed (but no example problems). No
Georgia Tech - MATH - 6338
MATH 6338HOMEWORK #2Due date: June 11, 2008Work the following problems and hand in your solutions. You may work together with other people in the class, but you must each write up your solutions independently. A subset of these will be selected