FAST '18 Birds-of-a-Feather Sessions

BoF Schedule

Lead or attend a BoF! Meet with your peers! Present new work! Don't miss these special activities designed to maximize the value of your time at the conference. Birds-of-a-Feather sessions are very informal gatherings of persons interested in a particular topic.

Vendor BoFs

Want to demonstrate a new product or discuss your company's latest technologies with FAST '18 attendees? Host a Vendor BoF! These sponsored one-hour sessions give companies a chance to talk about products and proprietary technology—and they include promotional benefits. Email if you're interested in sponsoring a Vendor BoF. More information about sponsorship opportunities is available here.

Scheduling a BoF

It's not too late! To schedule a BoF, simply write the BoF title as well as your name and affiliation on the BoF board located in the registration area. If you have a description of your BoF you'd like posted on this Web page, please schedule your BoF on the BoF board, then send its title, the organizer's name and affiliation, and the date, time, and location of the BoF to with "FAST '18 BoF" in the subject line.

Monday, February 12
ROOM CAPACITY 7:00 pm–8:00 pm 8:00 pm–9:00 pm 9:00 pm–10:00 pm 10:00 pm–11:00 pm
OCC Room 208 94 USENIX Women in Advanced Computing (WiAC) BoF Students and Young Professionals Meetup Board Game Night
Tuesday, February 13
ROOM CAPACITY 7:30 pm–8:30 pm 8:30 pm–9:30 pm 9:30 pm–10:30 pm
OCC Room 201 (No A/V) 26 Ceph Developer Meet & Greet GLUSTER: Extensible Distributed File System
OCC Room 208 94 NetApp Vendor BoF: Everything You Wanted to Know about WAFL but Were Afraid to Ask Current State of Scale Out File Systems in the Cloud
Hall East 550
Wednesday, February 14
ROOM CAPACITY 8:00 pm–9:00 pm 9:00 pm–10:00 pm 10:00 pm–11:00 pm
OCC Room 201 (No A/V) 26
OCC Room 208 94 Alibaba Vendor BoF:
1) Introduction to Alibaba Pangu Distributed System
2) Alibaba Cloud File System at Hyperscale
Hall East 550 Intro to Popper: Reproducible Evaluation of Systems Experiments

BoF Descriptions

USENIX Women in Advanced Computing (WiAC BoF)
Monday, February 12, 7:00 pm–8:00 pm, OCC Room 208
Let’s talk about women in advanced computing. All registered attendees—of all genders—are welcome to attend this BoF.

Students and Young Professionals Meetup
Monday, February 12, 8:00 pm–9:00 pm, OCC Room 208
Come for the refreshments, stay for the opportunity to meet and network with other students and young professionals.

Board Game Night
Monday, February 12, 9:00 pm–11:00 pm, OCC Room 208
Join us for some good old-fashioned board games. We’ll have some on hand, but bring your own games, too!

Ceph Developer Meet & Greet
Tuesday, February 13, 7:30 pm–8:30, OCC Room 201
Organizers: Greg Farnum and Sage Weil, Red Hat
Meet Ceph developers to ask your questions and hear a brief update on current topics in the Ceph community, including new programming models and disk storage plans. Users, developers, and hecklers are all welcome!

NetApp Vendor BoF: Everything You Wanted to Know about WAFL but Were Afraid to Ask
Tuesday, February 13, 7:30 pm–8:30, OCC Room 208
Speaker: Ram Kesavan
Share a beer and enjoy a discussion of NetApp's WAFL file system, including its design, features, and 25-year evolution. Bring your questions for host Ram Kesavan, a long time WAFL developer and author of several recent WAFL papers. Prizes for entertaining and insightful questions.

Current State of Scale Out File Systems in the Cloud
Tuesday, February 13, 8:30 pm–9:30, OCC Room 208

Dean Hildebrand, Google, and Sage Weil, Red Hat
This session will discuss the current state of scale-out file systems in the cloud. Are they on the way up? On the way down? Are their major obstacles to growing the set of applications and use cases that make use of scale-out file systems? What use cases should scale out file systems target? Come one, come all and voice your opinion!

Alibaba Vendor BoF
Wednesday, February 14, 8:00 pm–9:00 pm, OCC Room 208
Hong Tang, Chief Architect, Alibaba Cloud

Pangu: A Hyper-scale Storage System for Alibaba Cloud
Yikang Xu, Principal Engineer, Alibaba Group
Hyperscale cloud storage providers need to serve a wide range of customer workloads, deal with unprecedented scale and hyper growth, innovate fast, improve resource efficiency and keep cost low. We will share some of our thoughts, lessons and challenges from Alibaba's storage infrastructure side to how to achieve these goals in developing a unified distributed storage layer-Pangu system-as cloud storage foundation during the past decade.

Alibaba Cloud File System at Hyperscale
Qingda Lu, Staff Software Engineer, Alibaba Group
We will give a brief introduction of Alibaba Cloud NAS, a family of scalable file storage services in Alibaba Cloud with multiple protocol support. We will also discuss the technical challenges we faced and the enhancements we are working on.

Alibaba Dual-mode Flash Storage for Hyperscale Infrastructure
Shu Li, Engineer of Director, Alibaba Group
The increasing proliferation AI, IoT and Cloud has led to broadly diversified workloads and user cases in hyperscale data center. The traditional approach of building software above the standard SSDs possesses finite improvements, resulting in the sub-optimal storage performance. This work presents Alibaba's R&D endeavor to pursue optimal storage performance utilizing software/hardware co-design. Besides the kernel version, user-space open channel storage stack is closely optimized with the practical scenarios. The exploited gains include access latency reduction, QoS improvement, IO determinism, write amplification reduction, and lifespan extension.

Intro to Popper: Reproducible Evaluation of Systems Experiments
Wednesday, February 14, 8:00 pm–9:00, Hall East
Ivo Jimenez and Carlos Maltzahn, UC Santa Cruz
Experiments in storage systems are notoriously hard to reproduce; building, deploying and testing a storage system with the goal of replicating performance behavior is a difficult task. In an ideal scenario, it should be easy to recreate the environment where a test originally executed, so that others can quickly replicate results. Popper is a convention and CLI tool for implementing testing pipelines following best DevOps practices. Popper pipelines encompass the end-to-end experimental workflow that is typically followed by researchers (building, deploying, executing and analyzing results) and are relatively easy to execute: clone a repository, define environment variables, and execute a bash script. This BoF will introduce Popper, walk through examples and open up a discussion on how can we create community-wide pipelines to run storage systems experiments.