CSCI 680: Parallel Computer Architecture, Spring 2017

General Information

• Time and Location: MW 3:30PM-4:50PM, McGL 002

• Instructor: Adwait Jog (Personal Website)

• Office hours: MW 5:00PM to 6:30PM or by appointment, McGL 111

• Email: adwait@cs.wm.edu
  

• Deadlines: Jan 27 (add/drop deadline) and Mar 17 (withdraw deadline)

• Final exam: May 1, Monday, 9 AM to 12:00 noon, McGL 002 (Final Exam Schedule)

Prerequisites

Students are expected to have a good understanding of the basic computer organization and design, especially of the topics covered by CS 424-524, Computer Architecture. Please talk to the instructor if you do not satisfy this requirement.

Course Materials

Most of the material will be based on the research papers. However, the following optional textbook is recommended: Computer Architecture, Fifth Edition: A Quantitative Approach

Course Description

This course provides an in-depth understanding of the fundamental principles and design trade-offs involved in designing modern parallel computing architectures​.​ ​We will also discuss a range of latest papers that have appeared in leading international journals and conferences​ to understand ​​​​research issues associated with parallel computing​ systems​. Students are expected to read a variety of ​research ​papers, critique them and present them in the front of the class. In addition, students are expected to complete a ​semester-long research project​, and take a final exam.

Grade Distribution

• (I) Paper critiques and Homeworks: 20%

• (II) In-Class Presentations: 20%

• (III) Semester-long Research Project: 40%

• (IV) Final Exam: 20%

Discussion Forum, Announcements, and Submissions

• We will extensively use Piazza for discussions. All announcements will also be made through Piazza.

• Lecture slides from the instructor will be uploaded on Piazza.

• Critique and Homework Submissions via hard-copy in the class.

• Final Project Report via Email to the Instructor.

• Final grade submission via Banner.

(I) Paper Critiques and Homeworks

All students are required to submit a detailed critique for each paper we discuss in the class. However, the student who presents the paper in-class is not allowed to submit the critique for that particular paper. Deadline for critique submission is one week from when the paper is discussed completely in the class. Please submit any TEN critiques.

Submission Format: Each critique should not exceed one-page and must consists of four sections: 1) paper summary (2-3 lines), 2) strengths (2-3 lines), 3) weaknesses (2-3 lines), and 4) detailed comments (rest of the page). More details are already discussed in class and associated slides are submitted to the box folder (shared with students).

In addition to critiques, instructor may assign some homeworks as well.

You are encouraged to discuss the papers and homeworks with each other but should write critiques on your own.

(II) In-Class Presentations

Each student will present a maximum of two papers throughout the semester. If you plan to audit the course, you are required to present at least one paper. When you present a paper, be prepared to answer a variety of questions asked by the instructor or other fellow students. The goal is to make class lively. A list of papers will be provided to students. They can choose from that list or come up with their own suggestions. Suggestions would need approval from the instructor.

I expect students to first present necessary background and then paper details. The remaining time will be for discussion driven by the fellow students and the instructor. After your presentation is complete, send your final slides to the class on Piazza.

During each student presentation, each student (except the presenter) needs to fill the student feedback form. The form is available here. Please make sure you have a hard copy of it for every student presentation.

(III) Semester-Long Research Project

Students are expected to perform a semester-long research project. All projects need to be approved by the instructor. Please contact the instructor early to brainstorm potential project ideas.

Project Timeline

• Phase 1 – Project Determination:; Please send an email to the instructor by the deadline containing: 1) Project Name (think of this as your paper/report title), 2) Problem Statement, 3) Expected Infrastructure Platform Required, 4) Possible Outcomes and Deliverables.
  

• Phase 2 – Determination of (more) concrete project goals and Infrastructure Setup; Please meet the instructor during office hours to discuss the status of your project.

• Phase 3 – Motivation Results; Please meet the instructor during office hours to discuss the status of your project.

• Phase 4 – Final Project Report; Please email your final project report in PDF format. Please use the IEEE CAL format. The PDF should have these sections: 1) Problem Statement, 2) Introduction, 3) Background and Related Work, 4) Motivation Results, 5) Implementation Details, 6) Infrastructure Details, 7) Final Results, and 8) Conclusions.

(IV) Final Exam

A comprehensive final exam will be conducted in the finals week (May 1, Monday, 9AM to noon in McGL 002). It will cover all the topics covered in the class, including the content covered by the discussed research papers.

General Areas for In-class Presentations and Projects

• Core Design

• Memory Systems

• Near-Data Computing

• Accelerators (e.g., GPUs)

• Approximate Computing

• Emerging Memory/Storage Technologies

• Mobile Architectures

• Data Center Architectures

• Hardware Security

• Reliability and Dependable Systems

• Emerging Architectures: Quantum and DNA Architectures

Reading List and Other Resources

• Complete Reading List can be found here (Google Doc Link). The list is currently under construction.
  
  
  Useful Simulators/Tools
  

• GEM5, A Full System CPU Simulator

• MARSSx86, A Full System CPU Simulator

• McPAT, A power, area, and timing modeling framework

• Sniper, A Parallel/Fast CPU Simulator

• zsim, Another Fast CPU Simulator

• GPGPU-Sim, A GPU Simulator (models NVIDIA-style GPUs). Also, look at GPU-Wattch, A GPU Power Model

• Multi2sim, A CPU/GPU Simulator (models AMD-style GPUs)

• CPU-GPU Simulator, A Trace/Execution Driven CPU-GPU Heterogeneous Architecture Simulator

• GemDroid, A Mobile Architecture Simulator (GEM5 + Android Emulator)

Semester Schedule (Will be updated regularly)

Readings

• P1: A DNA-Based Archival Storage System, ASPLOS 2016

• P2: DNA-based Molecular Architecture with Spatially Localized Components, ISCA 2013

• P3: A Split Cache Hierarchy for Enabling Data-oriented Optimizations, HPCA 2017

• P4: Pilot Register File: Energy Efficient Register File for GPUs, HPCA 2017

• P5: TABLA: A Unified Template-based Framework for Accelerating Statistical Machine Learning, HPCA 2016

• P6: vDNN: Virtualized Deep Neural Networks for Scalable, Memory-Efficient Neural Network Design, MICRO 2016

• P7: Towards Pervasive and User Satisfactory CNN across GPU Microarchitectures, HPCA 2017

• P8: EIE: Efficient Inference Engine on Compressed Deep Neural Network, ISCA 2016

• P9: Phase-Aware Optimization in Approximate Computing, CGO 2017

• P10: Understanding and Optimizing Power Consumption in Memory Networks, HPCA 2017

• P11: GPGPU Performance and Power Estimation Using Machine Learning, HPCA 2015

• P12: Dynamic GPGPU Power Management using Adaptive Model Predictive Control, , HPCA 2017

• P13: Secure Dynamic Memory Scheduling Against Timing Channel Attacks, HPCA 2017

• P14: Reliability-Aware Scheduling on Heterogeneous Multicore Processors, HPCA 2017

• P15: Near-Optimal Access Partitioning for Memory Hierarchies with Multiple Heterogeneous Bandwidth Sources, HPCA 2017

• P16: Compute Caches, HPCA 2017

• P17: Architecting an Energy Efficient DRAM System for GPUs, HPCA 2017

• P18: Design and Analysis of an APU for Exascale Computing, HPCA 2017

• P19: G-Scalar: Cost-Effective Generalized Scalar Execution Architecture for Power-Efficient GPUs, HPCA 2017

• P20: Pagoda: Fine-Grained GPU Resource Virtualization for Narrow Tasks, PPoPP 2017

• P21: Power Attack Defense: Securing Battery-Backed Data Centers, ISCA 2016

• P22: Camouflage: Memory Traffic Shaping to Mitigate Timing Attacks, HPCA 2017

Academic Accommodations

It is the policy of The College of William and Mary to accommodate students with disabilities and qualifying diagnosed conditions in accordance with federal and state laws. Any student who feels s/he may need an accommodation based on the impact of a learning, psychiatric, physical, or chronic health diagnosis should contact Student Accessibility Services staff at 757-221-2509 or at sas@wm.edu to determine if accommodations are warranted and to obtain an official letter of accommodation. For more information, please click here.

Honor Code

Students are required to follow the Honor System of the College of William and Mary.