PARALLEL DATA LAB 

24th Annual Workshop & Retreat
October 17 - 19, 2016

INFO & Agenda

Retreat Agenda


Monday, October 17: CMU/Bedford Springs Resort
8:30 am - 9:00 am Registration & Continental Breakfast
-- SINGLETON ROOM, ROBERTS HALL
9:00 am - 9:30 am Welcome -- SINGLETON ROOM

Greg Ganger - Parallel Data Lab 2016 Retreat Welcome & Overview

9:30 am - 10:30 am Speaker -- SINGLETON ROOM

Kayvon Fatahalian & David Andersen – VizCloud ISTC: New Research Center on Cloud Systems for Visual Computing

10:30 am - 11:30 am Industry Poster Session -- ROBERTS HALL ATRIUM
11:30 am - 1:45 pm Bus Ride to Bedford Springs & Lunch -- FREW STREET
2:00 pm - 3:30 pm Session I: Specializing Storage Software
-- COLONNADE BALLROOM

Anuj Kalia – Designing High Performance RDMA Systems: Guidelines and Distributed Transactions

Abutalib Aghayev – Evolving ext4 for SMR Disks

3:30 pm - 3:45 pm Break -- CONSTITUTION HALL
3:45 pm - 6:15 pm Session II: Making it Easier to Use ML
-- COLONNADE BALLROOM

Kevin Hseih - Gaia – Geo-Distributed Machine Learning Approaching LAN Speeds

Henggang Cui – System Support for Automatic Machine Learning Tuning

Jinliang Wei – Automatic Parallelization for Efficient Machine Learning Training

Lin Ma – The Self-Driving DBMS

Michael Zhang – Avoiding Transaction Conflict via Statistical Scheduling

George Amvrosiadis – File Systems in the Era of Exascale Computing and Tiered Storage

6:15 pm - 7:30 pm Dinner -- EISENHOWER BALLROOM
7:30 pm - 8:30 pm Session III: Short Work-in-Progress Talks
-- COLONNADE BALLROOM

Qing Zheng – DeltaFS: Your Next HPC “/scratch” File System

Saurabh Kadekodi – Aging Gracefully with Geriatrix: A File System Aging Suite

Matt Perron – Write-Behind Logging

Pengtao Xie – Lighter-Communication Distributed Machine Learning via Sufficient Factor Broadcasting

8:30 pm - 10:30 pm Reception/Poster Session I -- COLONNADE BALLROOM

Tuesday, October 18: Bedford Springs Resort

8:15 am - 9:00 am Full Breakfast -- CRYSTAL DINING ROOM
Continental Breakfast
-- CONSTITUTION HALL
9:00 am - 10:30 am Session IV: Shared Storage and ML in the Cloud
-- COLONNADE BALLROOM

Timmy Zhu – SNC-Meister: Admitting More Tenants with Tail Latency SLOs

Aaron Harlap – Agile Elasticity in ML: How To Do it and How to Take Advantage of It

10:30 am - 10:45 am Break -- CONSTITUTION HALL
10:45 am - 12:15 pm Session V: Doing More with Memory
-- COLONNADE BALLROOM

Huanchen Zhang – Reducing the Storage Overhead of Main-memory OLTP Databases with Hybrid Indexes

Rajat Kateja – BBRAM++: Low-Cost Battery Backed DRAM

12:15 pm - 1:30 pm Lunch -- EISENHOWER BALLROOM
1:30 pm - 3:00 pm Socializing and Chat-filled Walk -- HOTEL GROUNDS
Wii games room available 1:00 pm to midnight --
REAGAN-B
3:00 pm - 4:30 pm Session VI: Correctness and Efficiency
-- COLONNADE BALLROOM

Ben Blum – Systematic Testing with Data-race Preemption Points

Yixin Luo – Improving SSD Lifetime with Flash Device Awareness

4:30 pm - 4:45 pm Break -- CONSTITUTION HALL
4:45 pm - 6:15 pm Session VII: Toward Better Scheduling and Analytics Performance
-- COLONNADE BALLROOM

Angela Jiang & Junwoo Park – JamaisVu: Robust Scheduling with Auto-estimated Runtimes

Jinliang Wei – Benchmarking Apache Spark with Machine Learning Applications

6:15 pm - 7:30 pm Dinner -- EISENHOWER BALLROOM
7:30 pm - 8:30 pm Session VIII: Short Work-In-Progress Talks
-- COLONNADE BALLROOM

Ziqi Wang – Implementing An In-Memory Concurrent Lock-Free B+Tree Index

Chris Fallin – Macro-scale Program Optimizations by Inferring Semantic Intent

Prashanth Menon – Optimizing Compiled Query Execution for Main-Memory Databases

Andy Pavlo – CMU DB App Catalog

8:30 pm - 10:00 pm Reception/Poster Session II -- COLONNADE BALLROOM

Wednesday, October 19: Bedford Springs Resort


Morning Workshop Talks
8:45am - 9:15am Full Breakfast -- CRYSTAL DINING ROOM
Continental Breakfast
-- CONSTITUTION HALL
9:15am - 11:30am Session IX: Storage, New Memory Models, and New ML Models
-- COLONNADE BALLROOM

George Amvrosiadis – Unobtrusive Storage Maintenance

Charlie McGuffey – Parallel Algorithms for Asymmetric Read-Write Costs

Jin Kyu Kim – STRADS: A Comprehensive Approach for Improving Performance and Programmability of Parallel ML

Afternoon Feedback, Departure
11:30am - 1:15pm Lunch and Industry Feedback
1:30pm Buses depart for Pittsburgh Airport & CMU (one in each direction)
4:00pm Arrive at Airport
4:00pm
Arrive at CMU


CONTACTS:


    Director, Parallel Data Lab
    VOICE: (412) 268-1297


    Executive Director, Parallel Data Lab
    VOICE: (412) 268-5485


    PDL Administrative Manager
    VOICE: (412) 268-6716