Preliminary Program Overview

Oct. 24, 2021, Sunday
16:00 — 20:00 Registration Desk Conference Reception
Oct. 25, 2021, Monday
7:00 — 16:00
Registration desk
08:30 — 9:00
 Opening Remarks
09:00 — 10:00
 Keynote 1
Intelligent Architectures for Intelligent Systems
Onur Mutlu, ETH Zurich
10:00 — 10:30
 Coffee Break
10:30 — 12:00
 Industry Session

World’s #1 CRM Scale Challenges(10:30)
Jason Ding, SVP Engineering, Salesforce

Making Enterprise SSD Smarter and Faster(11:00)
Yafei Yang CEO, Shenzhen Dapu Microelectronics Co., Ltd

Machine Reasoning – Improving AIOps for Intent Based Networks(11:30)
Pradeep Kathail, CTO Cisco
12:00 — 13:30
 Lunch Break
13:30 — 15:00
Session 1: Architecture
Chair: Jiayi Huang

Cache Compression with Efficient in-SRAM(13:30)
Xiaowei Wang (University of Michigan), Charles Augustine (Intel Corporation), Eriko Nurvitadhi (Intel Corporation), Ravi Iyer (Intel Corporation), Li Zhao (Alibaba Group) and Reetuparna Das (University of Michigan)

GO: Out-Of-Core Partitioning of Large Irregular Graphs(13:50)
Gurneet Kaur (University of California, Riverside) and Rajiv Gupta (University of California, Riverside)

PowerPrep: A power management proposal for user-facing datacenter workloads(14:10)
Vineetha Govindaraj (Pennsylvania State University), Sumitha George (North Dakota State University), Mahmut Kandemir (Pennsylvania State University), John Sampson (Pennsylvania State University) and Vijaykrishnan Narayanan (Pennsylvania State University)

Efficient NVM Crash Consistency by Mitigating Resource Contention(14:30)
Zhiyuan Lu (Michigan Technological University), Jianhui Yue (Michigan Technological University), Yifu Deng (Michigan Technological University) and Yifeng Zhu (University of Maine)

Migrating Software from x86 to ARM Architecture: An Instruction Prediction Approach(14:45)
Zong (Texas State University), Apan Qasem (Texas State University) and Jelena Tešić (Texas State University) 		Session 2: Storage
Chair: Tao Lu

NVSwap: Latency-Aware Paging using Non-Volatile Main Memory(13:30)
Yekang Wu (Washington State University Vancouver) and Xuechen Zhang (Washington State University Vancouver)

Edges: Evenly Distributing Garbage-Collections for Enterprise SSDs via Stochastic Optimization(13:50)
Shuyi Pei (University of Rhode Island), Qing Yang (University of Rhode Island) and Bin Li (University of Rhode Island)

Constant Time Garbage Collection in SSDs(14:10)
Reza Salkhordeh (Johannes Gutenberg University Mainz, Germany), Kevin Kremer (Johannes Gutenberg University Mainz, Germany), Lars Nagel (Loughborough University, UK), Dennis Maisenbacher (Johannes Gutenberg University Mainz, Germany), Hans Holmberg (Western Digital, Inc), Matias Bjørling (Western Digital, Inc) and André Brinkmann (Johannes Gutenberg University Mainz, Germany)

A Comprehensive Empirical Study of File Systems on Optane Persistent Memory(14:30)
Yang Yang (Huazhong University of Science and Technology) and Qiang Cao (Huazhong University of Science and Technology)

New Processing In-Storage Accelerator for Text Searching(14:45)
Yunxin Huang (Southeast University), Aiguo Song (Southeast University) and Yafei Yang (Shenzhen DAPU Microelectronics)
15:00 — 15:30
Coffee Break
15:30 — 17:30
Session 3: Network
Chair: Dipak Ghosal

Minimizing the Number of Rules to Mitigate Link Congestion in SDN-based Datacenters(15:30)
Rajorshi Biswas (Penn State Berks), Jie Wu (Temple University) and Yang Chen (Temple University)

An Incast-Coflow-Aware Minimum-Rate-Guaranteed Congestion Control Protocol for Datacenter Applications(15:50)
Zhijun Wang (The University of Texas at Arlington), Yunxiang Wu (Purple Mountain Laboratories), Rosenkrantz Stoddard (The University of Texas at Arlington), Ning Li (The University of Texas at Arlington), Minh Nguyen (The University of Texas at Arlington) and Hao Che (The University of Texas at Arlington)

Balancing Latency and Quality in Web Search(16:05)
Liang Zhou (University of California Riverside) and K. K. Ramakrishnan (University of California Riverside)
Design of A Real-Time Reconfigurable Multi-Path Traffic Mirroring System(16:20)
Liang-Min Wang (Intel Corp), Timothy Miskell (Intel Corp), Edwin Verplanke (Intel Corp) and John Morgan (Intel Corp)

Terrestrial and Space-based Cloud Computing with Scalable, Responsible and Explainable Artificial Intelligence - A Position Paper(16:35)
Davide Martizzi (Albeado Inc), Partha Datta Ray (Albeado Inc)

Implementation of a high-throughput virtual switch port mirroring system(16:50)
Liang-Min Wang (Intel Corp), Timothy Miskell (Intel Corp), Patrick Fu (Intel Corp), Cunming Liang (Intel Corp) and Edwin Verplanke (Intel Corp)

 Session 4: Architecture
Chair: Daniel Wong

Vulkan vs OpenGL ES: Performance and Energy Efficiency Comparison on the big(15:30)
Michael Lujan (Texas State University), Michael McCrary (Texas State University), Blake W. Ford (Texas State University), and Ziliang Zong (Texas State University)

LocalityGuru: A PTX Analyzer for Extracting Thread Block-level Locality in GPGPUs(15:50)
Devashree Tripathy (University of California, Riverside), Amirali Abdolrashidi (University of California, Riverside), Quan Fan (University of California, Riverside), Daniel Wong (University of California, Riverside) and Manoranjan Satpathy (Indian Institute of Technology Bhubaneswar)

GPU-Assisted Memory Expansion(16:10)
Pisacha Srinuan (University of Louisiana at Lafayett), Purushottam Sigdel (University of Louisiana at Lafayett), Xu Yuan (University of Louisiana at Lafayett), Lu Peng (Louisiana State University), Paul Darby (University of Louisiana at Lafayett), Christopher Aucoin (University of Louisiana at Lafayett) and Nian-Feng Tzeng (University of Louisiana at Lafayett)

E2E Visual Analytics: Achieving >10X Edge/Cloud Optimizations(16:30)
Chaunte Lacewell (Intel Labs), Nilesh Ahuja (Intel Labs), Pablo Munoz (Intel Labs), Parual Datta (Intel Labs), Ragaad Altarawneh (Intel Labs), Vui Seng Chua (Intel Labs), Nilesh Jain (Intel Labs), Omesh Tickoo (Intel Labs) and Ravi Iyer (Intel Labs)

Locality-aware Thread Block Design in Single and Multi-GPU Graph Processing(16:45)
Quan Fan (University of California, Riverside) and Zizhong Chen (University of California, Riverside

MLPP: Exploring Transfer Learning and Model Distillation for Predicting Application Performance(17:00)
Jashwant Raj Gunasekaran (The Pennsylvania State University) and Cyan Subhra Mishra (The Pennsylvania State University)
17:30 — 18:00
 Social
18:00 — 20:00
Inception Dinner
Oct. 26, 2021, Tuesday
09:00 — 10:00
 Keynote 2
Vulnerabilities in Federated Learning Networks
Prasant Mohapatra, UC Davis
10:00 — 10:30
 Coffee Break
10:30 — 12:00
Session 5: Architecture
Chair: Nian-Feng Tzeng

ICAP: Designing Inrush Current Aware Power Gating Switch for GPGPU(10:30)
Hadi Zamani (University of California, Riverside), Devashree Tripathy (University of California, Riverside), Ali Jahanshahi (University of California, Riverside) and Daniel Wong (University of California, Riverside)

Machine Learning-based Vulnerability Study of Interpose PUFs as Security Primitives for Resource-Constrained IoT Networks(10:50)
Bipana Thapaliya (Texas Tech University), Khalid Mursi (University of Jeddah, Jeddah) and Yu Zhuang (Texas Tech University)

Reducing the Training Overhead of the HPC Compression Autoencoder via Dataset Proportioning(11:10)
Tong Liu (Temple University), Shakeel Alibhai (Temple University), Jinzhen Wang (New Jersey Institute of Technology), Qing Liu (New Jersey Institute of Technology) and Xubin He (Temple University, Philadelphia)

Exploring Storage Device Characteristics of A RISC-V Little-core SoC(11:30)
Tao Lu (Marvell Technology Group)

Deflection-aware Routing Algorithm in Network on Chip against Soft Errors and Crosstalk Faults(11:45)
Hadi Zamani (University of California, Riverside), Zahra Shirmohammadi (Shahid Rajaee Teacher Training University) and Ali Jahanshahi (University of California, Riverside) 		Session 6: Storage
Chair: Xuechen Zhang

Improving Relational Database Upon the Arrival of Storage Hardware with Built-in Transparent Compression(10:30)
Yifan Qiao (Rensselaer Polytechnic Institute), Xubin Chen (Rensselaer Polytechnic Institute), Jingpeng Hao (Rensselaer Polytechnic Institute), Jiangpeng Li (ScaleFlux Inc.), Qi Wu (ScaleFlux Inc.), Jingqiang Wang (ScaleFlux Inc.), Yang Liu (ScaleFlux Inc.) and Tong Zhang (Rensselaer Polytechnic Institute)


Cached Mapping Table Prefetching for Random Reads in Solid-State Drives(11:50)
Xiaojun Ruan (California State University, East Bay), Xunfei Jiang (California State University, Northridge) and Haiquan Chen (California State University, Sacramento)

A Content-Aware Learning Cache for Storage Systems(11:10)
Maher Kachmar (Northeastern University) and David Kaeli (Northeastern University)

On Adapting the Cache Block Size in SSD Caches(11:30)
Nikolaus Jeremic (University of Rostock, Germany), Helge Parzyjegla (University of Rostock, Germany), and Gero Mühl (University of Rostock, Germany)
EFLOG: A Full Streaming Logging scheme for Erasure Coding in Cloud Storage Systems(11:45)
Lei Sun (Huazhong University of Science and Technology), Shucheng Wang (Huazhong University of Science and Technology), Qiang Cao (Huazhong University of Science and Technology) and Changsheng Xie (Huazhong University of Science and Technology)
12:00 — 13:30
Lunch Break
13:30 — 14:30
Keynote 3
QoS for Edge/Cloud Architectures: Past, Present and Future
Ravi Iyer, Intel
14:30 — 15:00
Coffee Break
15:00 — 17:00
Session 7: Network
Chair: TBA

Decoupling Control and Data Transmission in RDMA Enabled Cloud Data Centers(15:00)
Qingyue Liu (Rice University) and Peter Varman (Rice University)

Leveraging Network Delay Variability to Improve QoE of Latency Critical Services(15:20)
Sambit Shukla (UC Davis) and Matthew Farrens (UC Davis)

Egress Peer Engineering over MPLS Label Unicast Transport(15:40)
Sundaram T Radhakrishnan (Cisco System, Inc.), and Satya R. Mohanty (Cisco System, Inc.)

A Congestion Aware Multi-Path Label Switching in Data Centers Using Programmable Switches(15:55)
Yeim-Kuan Chang (National Cheng Kung University), Hung-Yen Wang (National Cheng Kung University) and Yu-Hsiang Lin (National Cheng Kung University)
Towards a Proactive Lightweight Serverless Edge Cloud for Internet-of-Things Applications(16:10)
Ian-Chin Wang (University of California, Riverside), Shixiong Qi (University of California, Riverside), Elizabeth Liri (University of California, Riverside) and K.K. Ramakrishnan (University of California, Riverside)

Towards Energy-Efficient and Real-Time Cloud Computing(16:25)
Taha Tekreeti (Auburn University), Tathagata Bhattacharya (Auburn University), Ting Cao (Auburn University), Xiaopu Peng (Auburn University), Jianzhou Mao (Auburn University), Xiao Qin (Auburn University) and Wei-Shinn Ku (Auburn University)

Session 8: Storage
Chair: Xunfei Jiang

Implementing Flash-Cached Storage Systems Using Computational Storage Drive with Built-in Transparent Compression(15:00)
Jingpeng Hao (Rensselaer Polytechnic Institute), Xubin Chen (Rensselaer Polytechnic Institute), Yifan Qiao (Rensselaer Polytechnic Institute), Yuyang Zhang (Rensselaer Polytechnic Institute) and Tong Zhang (Rensselaer Polytechnic Institute)

PLMC: A Predictable Tail Latency Mode Coordinator for Shared NVMe SSD with Multiple Hosts(15:20)
Tanaya Roy (Temple University), Jit Gupta (Temple University), Krishna Kant (Temple University), Amitangshu Pal (Indian Institute of Technology Kanpur), Dave Minturn (Intel Corp) and Arash Tavakkol (Fortum, Switzerland)

Characterizing AI Model Inference Applications Running in SGX Environment(15:40)
Shixiong Jing (The Pennsylvania State University), Qinkun Bao (The Pennsylvania State University), Pei Wang (Independent), Xulong Tang (University of Pittsburgh) and Dinghao Wu (The Pennsylvania State University)

Flow Scheduling in a Heterogeneous NFV Environment using Reinforcement Learning(16:00)
Chun-Jen Lin (University of Massachusetts Lowell), Yan Luo (University of Massachusetts Lowell), Liang-Min Wang (Intel Corporation) and Li-De Chen (Intel Corporation)

A Case Study of Migrating RocksDB on Intel Optane Persistent Memory(16:15)
Ziyi Lu (Huazhong University of Science and Technology) and Qiang Cao (Huazhong University of Science and Technology)

Fast Variable-Grained Resemblance Data Deduplication For Cloud Storage(15:35)
Xuming Ye (University of South China), Jia Tang (University of South China), Wenlong Tian (University of South China), Zhiyong Xu (Suffolk University), Ruixuan Li (Huazhong University of Science and Technology), Weijun Xiao (Virginia Commonwealth University) and Yuqing Geng (University of South China)
17:00 — 18:00
 Social
18:00 — 20:00
Banquet