Main Program

Saturday, June 18 – Workshops and Tutorials

8:00am-9:00am

Breakfast
(for Workshop and Tutorial attendees only)

9:00am-12:30pm***

AM Workshops and Tutorials

12:30pm-2:00pm

Lunch
(for Workshop and Tutorial attendees only)

2:00pm-5:30pm

PM Workshops and Tutorials

***Please check the start/end times of the individual workshops and tutorials. Some may start earlier or later than the default time of 9:00am.

Sunday, June 19 – Workshops and Tutorials

8:00am-9:00am

Breakfast
(for Workshop and Tutorial attendees only)

9:00am-12:30pm

AM Workshops and Tutorials

12:30pm-2:00pm

Lunch
(for Workshop and Tutorial attendees only)

2:00pm-5:30pm

PM Workshops and Tutorials

 

5:30pm-7:30pm

Welcome Reception (for all registered attendees)

Monday, June 20 – Start of Main Symposium

7:30am

Breakfast

 

8:15am

Welcome from General and Program Chairs

8:30am-9:30am

Keynote 1

Chair: Gabriel Loh

Doug Carmean (Microsoft): “Quantum and Cryo and DNA, oh my! Sights Along the New Yellow Brick Road”

9:30am-10:30am

Lightning Talks 1 [slides]

Chair: André Seznec

10:30am-11:00am

Coffee Break

 

 

11:00am-12:00pm

Session 1A: Neural Networks 1

Chair: Michael Taylor
UCSD

Cnvlutin: Ineffectual-Neuron-Free Deep Convolutional Neural Network Computing [slides]

Jorge Albericio

University of Toronto

Patrick Judd

Tayler Hetherington

University of British Columbia

Tor Aamodt

Natalie Enright Jerger

University of Toronto

Andreas Moshovos

ISAAC: A Convolutional Neural Network Accelerator with In-Situ Analog Arithmetic in Crossbars [slides]

Ali Shafiee

University
of Utah

Anirban Nag

Naveen Muralimanohar

HP Labs

Rajeev Balasubramonian

University of Utah

John Paul Strachan

HP Labs

Miao Hu

R. Stanley Williams

Vivek Srikumar

University of Utah

A Novel Processing-in-memory Architecture for Neural Network Computation in ReRAM-based Main Memory [slides]

Ping Chi

UC Santa Barbara

Shuangchen Li

Cong Xu

HP Labs

Tao
Zhang

Nvidia

Jishen Zhao

UC
Santa Cruz

Yongpan Liu

Tsinghua University

Yu Wang

Yuan Xie

UC Santa Barbara

11:00am-12:00pm

Session 1B: Heterogeneous Architecture / Approximate Computing

Chair: Murali Annavaram
USC

Asymmetry-Aware Work-Stealing Runtimes [slides]

Christopher Torng

Cornell University

Moyang Wang

Christopher Batten

Morpheus: Creating Application Objects Efficiently for Heterogeneous Computing [slides]

Hung-Wei Tseng

University of California, San Diego

Qianchen Zhao

Yuxiao Zhou

Mark Gahagan

Steven Swanson

Towards Statistical Guarantees in Controlling Quality Tradeoffs in Approximate Acceleration [slides]

Divya Mahajan

Georgia Institute of Technology

Amir Yazdanbakhsh

Jongse Park

Bradley Thwaites

Hadi Esmaeilzadeh

 

12:00pm-1:30pm

Lunch

 

1:30pm-2:30pm

Session 2A: Caches

Chair: Carole Wu
ASU

Back to the Future: Leveraging Belady’s Algorithm for Improved Cache Replacement [slides]

Akanksha Jain

University of Texas at Austin

Calvin Lin

Efficient Synonym Filtering and Scalable Delayed Translation for Hybrid Virtual Caching [slides]

Chang Hyun Park

KAIST

Taekyung Heo

Jaehyuk Huh

LAP: Loop-Block Aware Inclusion Properties for Energy-Efficient Asymmetric Last Level Caches [slides]

Hsiang-Yun Cheng

Penn State / UCSB

Jishen Zhao

UCSC

Jack Sampson

Penn State

Mary Jane Irwin

Aamer Jaleel

Nvidia

Yu Lu

Qualcomm

Yuan Xie

UCSB

1:30pm-2:30pm

Session 2B: Hardware Design

Chair: Mattan Erez
UT Austin

Automatic Generation of Efficient Accelerators for Reconfigurable Hardware [slides]

Raghu Prabhakar

Stanford University

David Koeplinger

Yaqi Zhang

Christina Delimitrou

Christos Kozyrakis

Kunle Olukotun

Strober: A Fast and Accurate Sample-Based Energy Simulation Framework for Arbitrary RTL [slides]

Donggyu Kim

UC Berkeley

Adam Izraelevitz

Christopher Celio

Hokeun Kim

Brian Zimmer

Yunsup Lee

Jonathan Bachrach

Krste Asanovic

PowerChop: Identifying and Managing Non-critical Units in Hybrid Processor Architectures [slides]

Michael A. Laurenzano

University of Michigan

Yunqi Zhang

Jiang Chen

Lingjia Tang

Jason Mars

2:30pm-3:30pm

Session 3A: Accelerators

Chair: Trey Cain
Qualcomm

Biscuit: A Framework for Near-Data Processing of Big Data Workloads [slides]

Boncheol Gu

Samsung Electronics Co.

Andre Sungho Yoon

Duck-Ho Bae

Insoon Jo

Jinyoung Lee

Jonghyun Yoon

Jeong Kang

Moonsang Kwon

Sangyeun Cho

Jaeheon Jeong

Duckhyun Chang

Energy Efficient Architecture for Graph Analytics Accelerators [slides]

Muhammet Mustafa Ozdal

Bilkent University

Serif Yesil

Taemin Kim

Intel Corporation

Andrey Ayupov

John Greth

Steven Burns

Ozcan Ozturk

Bilkent University

ASIC Clouds: Specializing the Datacenter [slides]

Ikuo Magaki

UC San Diego, Toshiba

Moein Khazraee

UC San Diego

Luis Vega

Michael Bedford Taylor

2:30pm-3:30pm

Session 3B: GPUs I

Chair: Chia-Lin Yang
NTU

APRES: Improving Cache Efficiency by Exploiting Load Characteristics on GPUs [slides]

Yunho Oh

Yonsei University

Keunsoo Kim

Myung Kuk Yoon

Jong Hyun Park

Yongjun Park

Hongik University

Won Woo Ro

Yonsei University

Murali Annavaram

University of Southern California

Transparent Offloading and Mapping (TOM): Enabling Programmer-Transparent Near-Data Processing in GPU Systems [slides]

Kevin Hsieh

CMU

Eiman Ebrahimi

Nvidia

Gwangsun
Kim

KAIST

Niladrish Chatterjee

Nvidia

Mike O’Connor

Nandita Vijaykumar

CMU

Onur Mutlu

Stephen W. Keckler

Nvidia

Warped-Slicer: Efficient Intra-SM Slicing through Dynamic Resource Partitioning for GPU Multiprogramming [slides]

Qiumin Xu

University of Southern California

Hyeran Jeon

Keunsoo Kim

Yonsei University

Won Woo Ro

Murali Annavaram

University of Southern California

 

3:30pm-4:00pm

Break

 

4:00pm-5:00pm

Session 4A: Neural Networks 2

Chair: Shubu Mukherjee
Cavium

EIE: Efficient Inference Engine on Compressed Deep Neural Network [slides]

Song Han

Stanford University

Xingyu Liu

Huizi Mao

Tsinghua University

Jing Pu

Stanford University

Ardavan Pedram

Mark Horowitz

Bill Dally

RedEye: Analog ConvNet Image Sensor Architecture for Continuous Mobile Vision [slides]

Robert LiKamWa

Rice University

Yunhui Hou

Mia Polansky

Yuan Gao

Lin Zhong

Minerva: Enabling Low-Power, High-Accuracy Deep Neural Network Accelerators [slides]

Brandon Reagen

Harvard University

Paul Whatmough

Robert Adolf

Saketh Rama

Hyunkwang Lee

Saekyu Lee

Jose Miguel Hernandez Lobato

Gu-Yeon Wei

David Brooks

4:00pm-5:00pm

Session 4B: NoC / Virtualization

Chair: David Wentzlaff
Princeton

Opportunistic Competition Overhead Reduction for Expediting Critical Section in NoC based CMPs [slides]

Yuan Yao

KTH Royal Institute of Technology, Sweden

Zhonghai Lu

Short-Circuit Dispatch: Accelerating Virtual Machine Interpreters on Embedded Processors [slides]

Channoh Kim

Sungkyunkwan University (SKKU)

Sungmin Kim

Hyeon Gyu Cho

Dooyoung Kim

Jaehyeok Kim

Younghwan Oh

Hakbeom Jang

Jae W. Lee

A Measurement Study of ARM Virtualization Performance [slides]

Christoffer Dall

Columbia University

Shih-Wei Li

Jintack Lim

Jason Nieh

Georgios Koloventzos

5:00pm-5:40pm

Session 5A: Cache / Memory Compression

Chair: Jung Ho Ahn
SNU

Base-Victim Compression: An Opportunistic Cache Compression Architecture [slides]

Jayesh Gaur

Intel

Alaa R. Alameldeen

Sreenivas Subramoney

Bit-Plane Compression: Transforming Data for Better Compression in Many-core Architectures [slides]

Jungrae Kim

The University of Texas at Austin

Michael Sullivan

Nvidia

Esha Choukse

The University of Texas at Austin

Mattan Erez

5:00pm-5:40pm

Session 5B: Reliability 1

Chair: Koji Inoue
Kyushu

XED: Exposing On-Die Error Detection Information for Strong Memory Reliability [slides]

Prashant J. Nair

Georgia Institute of Technology

Vilas Sridharan

Advanced Micro Devices, Inc.

Moinuddin K. Qureshi

Georgia Institute of Technology

Production-Run Software Failure Diagnosis via Adaptive Communication Tracking [slides]

Mohammad Mejbah ul Alam

University of Texas at San Antonio

Abdullah Muzahid

 

5:45pm-7:15pm

Business Meeting

Tuesday, June 21

7:30am

Breakfast

8:30am-9:30am

Keynote 2

Chair: Sang Lyul Min
Seok-Hee Lee (SK Hynix): “Memory’s value positioning in servers: when Moore’s Law sees the end of the road”

9:30am-10:30am

Lightning Talks 2 [slides]

Chair: André Seznec

10:30am-11:00am

Break

 

11:00am-12:00pm

Session 6: Neural Networks 3

Chair: Emre Ozer
ARM

Eyeriss: A Spatial Architecture for Energy-Efficient Dataflow for Convolutional Neural Networks [slides]

Yu-Hsin Chen

MIT

Joel Emer

MIT/NVIDIA

Vivienne Sze

MIT

Neurocube: A Programmable Digital Neuromorphic Architecture with High-Density 3D Memory [slides]

Duckhwan Kim

Georgia Institute of Technology

Jaeha Kung

Sek Chai

SRI International

Sudhakar Yalamanchili

Georgia Institute of Technology

Saibal Mukhopadhyay

Cambricon: An Instruction Set Architecture for Neural Networks [slides]

Shaoli Liu

Chinese Academy of Sciences

Zidong Du

Jinhua Tao

Dong Han

Tao Luo

Yuan Xie

UCSB

Yunji Chen

Chinese Academy of Sciences

Tianshi Chen

 

12:00pm-2:00pm

Awards Lunch

2:00pm-2:30pm

Room Setup/Reconfiguration

 

2:30pm-3:50pm

Session 7A: Microarchitecture

Chair: Antonio Gonzalez
UPC

Decoupling Loads for Nano-Instruction Set Computers [slides]

Ziqiang Huang

Duke Univeristy

Andrew Hilton

Benjamin Lee

Future Vector Microprocessor Extensions for Data Aggregations [slides]

Timothy Hayes

Barcelona Supercomputing Center

Oscar Palomar

Osman Unsal

Adrian Cristal

Mateo Valero

Efficiently Scaling Out-of-Order Cores for Simultaneous Multithreading [slides]

Faissal Sleiman

U. Michigan / Pure Storage

Thomas Wenisch

U. Michigan

Accelerating Dependent Cache Misses with an Enhanced Memory Controller [slides]

Milad Hashemi

UT Austin

Khubaib

Apple

Eiman Ebrahimi

Nvidia

Onur Mutlu

CMU

Yale Patt

UT Austin

2:30pm-3:50pm

Session 7B: Datacenters

Chair: Sandhya Dwarkadas
Rochester

Treadmill: Attributing the Source of Tail Latency through Precise Load Testing and Statistical Inference [slides]

Yunqi Zhang

University of Michigan

David Meisner

Facebook Inc.

Jason Mars

University of Michigan

Lingjia Tang

Dynamo: a Data Center-Wide Power Management System [slides]

Qiang Wu

Facebook Inc.

Qingyuan Deng

Bin Li

Lakshmi Ganesh

Justin Meza

Chang-Hong Hsu

Facebook Inc. / Univ. of Michigan Ann Arbor

Yun Jin

Facebook Inc.

Yee Jiun Song

Sanjeev Kumar

Peak Efficiency Aware Scheduling for Highly Energy Proportional Servers [slides]

Daniel Wong

University of California, Riverside

Power Attack Defense: Securing Battery-Backed Data Centers [slides]

Chao Li

Shanghai Jiao Tong University

Zhenhua Wang

Xiaofeng Hou

Haopeng Chen

Xiaoyao Liang

Minyi Guo

 

4:00pm-6:00pm

Excursion

6:00pm

Banquet

Wednesday, June 22

8:00am

Breakfast

 

9:00am-10:00am

Session 8A: Memory 1

Chair: John Kim
KAIST

DRAF: A Low-Power DRAM-Based Reconfigurable Acceleration Fabric [slides]

Mingyu Gao

Stanford University

Christina Delimitrou

Dimin Niu

Samsung Semiconductor Inc.

Krishna T. Malladi

Hongzhong Zheng

Bob Brennan

Christos Kozyrakis

Stanford University

Mellow Writes: Extending Lifetime in Resistive Memories through Selective Slow Write Backs [slides]

Lunkai Zhang

UC Santa Barbara

Brian Neely

Diana Franklin

University of Chicago

Dmitri Strukov

UC Santa Barbara

Yuan Xie

Fred Chong

University of Chicago

MITTS: Memory Inter-arrival Time Traffic Shaping [slides]

Yanqi Zhou

Princeton University

David Wentzlaff

9:00am-10:00am

Session 8B: Emerging Architectures

Chair: Alvin Lebeck
Duke

The Anytime Automaton [slides]

Joshua San Miguel

University of Toronto

Natalie Enright Jerger

Accelerating Markov Random Field Inference using Molecular Optical Gibbs Sampling Units [slides]

Siyang Wang

Duke University

Xiangyu Zhang

Yuxuan Li

Ramin Bashizade

Song Yang

Chris Dwyer

Alvin R. Lebeck

Evaluation of an Analog Accelerator [slides]

Yipeng Huang

Columbia University

Ning Guo

Mingoo Seok

Yannis Tsividis

Simha Sethumadhavan

 

10:00am-10:30am

Break

 

10:30am-11:30am

Session 9A: GPUs 2

Chair: Kei Hiraki
U Tokyo

LaPerm: Locality Aware Scheduler for Dynamic Parallelism on GPUs [slides]

Jin Wang

Georgia Institute of Technology

Norm Rubin

NVIDIA Research

Albert Sidelnik

Sudhakar Yalamanchili

Georgia Institute of Technology

ActivePointers: Software Address Translation Layer on GPUs [slides]

Sagi Shahar

Technion

Shai Bergman

Mark Silberstein

Virtual Thread: Maximizing Thread-Level Parallelism beyond GPU Scheduling Limit [slides]

Myung Kuk Yoon

Yonsei University

Keunsoo Kim

Sangpil Lee

Won Woo Ro

Murali Annavaram

University of Southern California

10:30am-11:30am

Session 9B: Reliability 2

Chair: Yoav Etsion
Technion

All-Inclusive ECC: Thorough End-to-End Protection for Reliable Computer Memory [slides]

Jungrae Kim

The University of Texas at Austin

Michael Sullivan

Nvidia

SangKug Lym

The University of Texas at Austin

Mattan Erez

Rescuing Uncorrectable Fault Patterns in On-chip Memories through Error Pattern Transformation [slides]

Henry Duwe

University of Illinois at Urbana-Champaign

Xun Jian

Daniel Petrisko

Rakesh Kumar

RelaxFault Memory Repair [slides]

Dong Wan Kim

The University of Texas at Austin

Mattan Erez

11:30am-12:30pm

Session 10A: Energy-efficient Computing

Chair: Gilles Pokam
Intel

Using Multiple Input, Multiple Output Formal Control to Maximize Resource Efficiency in Architectures [slides]

Raghavendra Pradyumna Pothukuchi

University of Illinois at Urbana-Champaign

Amin Ansari

Qualcomm Research

Petros G. Voulgaris

University of Illinois at Urbana-Champaign

Josep Torrellas

Exploiting Dynamic Timing Slack for Energy Efficiency in Ultra-Low-Power Embedded Systems [slides]

Hari Cherupalli

University of Minnesota

Rakesh Kumar

University of Illinois at Urbana-Champaign

John Sartori

University of Minnesota

CASH: Supporting IaaS Customers with a Sub-core Configurable Architecture [slides]

Yanqi Zhou

Princeton University

Henry Hoffmann

University of Chicago

David Wentzlaff

Princeton University

11:30am-12:30pm

Session 10B: Memory 2

Chair: Moinuddin Qureshi
GaTech

Boosting Access Parallelism to PCM-Based Main Memory [slides]

Mohammad Arjomand

Pennsylvania State University

Mahmut T. Kandemir

Anand Sivasubramaniam

Chita Das

Agile Paging: Exceeding the Best of Nested and Shadow Paging [slides]

Jayneel Gandhi

University of Wisconsin – Madison

Mark D. Hill

Michael M. Swift

Energy Efficient Data Encoding in DRAM Data Channels exploiting Data Value Similarity [slides]

Hoseok Seol

KAIST

Wongyu Shin

Jaemin Jang

Jungwhan Choi

Jinwoong Suh

Lee-Sup Kim

 

12:30pm

Conference Closing