Sponsored By



In Cooperation With

Supporters

Gold Sponsors

SK hynix

Silver Sponsors

Google

DELL Technologies

Follow us



Logo design by Michal O. Zadok.



Full Program

Day 1: Monday, July 8th, 2024

12:30 - 13:00 PDT Coffee Break
13:00 - 13:15 PDT Opening Remarks
13:15 - 14:30 PDT Session 1: Data Encoding and Compression
14:30 - 14:50 PDT Coffee Break
14:50 - 15:40 PDT Session 2: LLMs to the rescue
15:40 - 16:30 PDT Panel: Storage and Machine Learning
16:30 - 17:20 PDT Session 3: Considering Workloads

Day 2: Tuesday, July 9th, 2024

08:30 - 09:00 PDT Coffee Break
09:00 - 10:00 PDT Keynote by Carlos Maltzahn
10:00 - 11:15 PDT Session 4: Keeping Data Correct
11:15 - 11:30 PDT Break
11:30 - 12:20 PDT Session 5: Minimizing Latencies
12:20 - 12:40 PDT Honoring Erik Riedel (We Will Always Miss You)
12:40 - 14:00 PDT Lunch Break (On Your Own)
14:00 - 15:15 PDT Session 6: Storage and Machine Learning
15:15 - 16:30 PDT Session 7: Storage Technologies
16:30 - 16:45 PDT Concluding Thoughts
16:45 - 17:20 PDT Coffee and Networking



Keynote Talks

The impact of Ceph on open source at UC Santa Cruz and beyond
Carlos Maltzahn
[Abstract], [Speaker Bio], [Slides]
Abstract:

The Ceph free and open source software-defined storage platform evolved from Sage Weil's working prototype file system for a 2005 summer project at UC Santa Cruz to IBM Storage Ceph, one of the most successful storage systems today. It combines openness with the support by many vendors and is a system of choice for national/international cyberinfrastructures for open science, including in storage systems research. In the presentation, I will briefly go over some of the lessons that enabled Ceph’s success and then focus on how that motivated the institutional support for open source software at UC Santa Cruz and beyond. With the help of Sage Weil, we established the Center for Research in Open Source Software (CROSS) in 2015 to support/promote open source research products and embed them in open source ecosystems. In the meantime an increasing number of major corporations established open source program offices (OSPOs) to implement their open source strategies and to manage strategic dependencies on open source software communities. Similar dependencies also exist in university research enterprises, but there the focus is on unlocking the potential of open source communities to amplify the impact of scientific research. Thanks to the success of CROSS and to the generous support from Sloan Foundation, in 2022 UC Santa Cruz became one of the first universities to establish an OSPO. And earlier this year, UC Santa Cruz is leading the first Network of OSPOs across six University of California campuses, including UC Berkeley, UC Davis, UCLA, UC San Diego, and UC Santa Barbara — again funded by Sloan Foundation. This Network focusses on (1) the discovery and measurement of university open source landscapes, (2) training, development, and mentoryship, and (3) open source project sustainability.

Bio:

Carlos Maltzahn retired from UC Santa Cruz on December 15, 2023. He was the founder and director of the UC Santa Cruz Center for Research in Open Source Software (CROSS) and the Open Source Program Office (OSPO) UC Santa Cruz. He also co-founded the Systems Research Lab, known for its cutting-edge work on programmable storage systems, big data storage & processing, scalable data management, distributed system performance management, and practical reproducible evaluation of computer systems. Carlos joined UC Santa Cruz in 2004, after five years at Netapp working on network-intermediaries and storage systems. In 2005 he co-founded and became a key mentor on Sage Weil’s Ceph project. In 2008 Carlos became a member of the computer science faculty at UC Santa Cruz and has graduated ten Ph.D. students since. Carlos graduated with a M.S. and Ph.D. in Computer Science from University of Colorado at Boulder.



Panels

Storage and Machine Learning - What can we learn from each other?
Assaf Eisenman (Meta), Zhichao Cao (ASU), Janki Bhimani (FSU), Erez Zadok (Stony Brook)

Sessions

Session 1: Data Encoding and Compression
Session Chairs: Amy Tai (Google)

Is Low Similarity Threshold A Bad Idea in Delta Compression?
[Paper ] [Slides]
Hongming Huang (City University of Hong Kong), Chun Jason Xue (Mohamed bin Zayed University of Artificial Intelligence), Nan Guan (City University of Hong Kong), Hong Xu (The Chinese University of Hong Kong)

Dictionary Based Cache Line Compression
[Paper]
Daniel Cohen, Sarel Cohen, Dalit Naor (The Academic College of Tel Aviv-Yaffo, Israel), Daniel Waddington (IBM Research), Moshik Hershcovitch (Tel-Aviv University & IBM Research)

Revisiting Erasure Codes: A Configuration Perspective
[Paper ] [Slides]
Runzhou Han, Chao Shi, Tabassum Mahmud (Iowa State University), Zeren Yang (University of Wisconsin-Madison), Vladislav Esaulov, Lipeng Wan (Georgia State University), Yong Chen, Jim Wayda, Matthew Wolf (Samsung), Mai Zheng (Iowa State University)


Session 2: LLMs to the rescue
Session Chairs: Sudarsun Kannan (Rutgers University)

Can Modern LLMs Tune and Configure LSM-based Key-Value Stores?
[Paper ] [Slides]
SK Hynix Best Paper Award Winner!
Viraj Thakkar, Madhumitha Sukumar, Jiaxin Dai, Kaushiki Singh, Zhichao Cao (Arizona State University)

ION: Navigating HPC I/O Optimization Journey using Large Language Models
[Paper ] [Slides]
Chris Egersdoerfer, Arnav Sareen (University of North Carolina at Charlotte), Jean Luca Bez (Lawrence Berkeley National Laboratory), Suren Byna (The Ohio State University), Dong Dai (University of North Carolina at Charlotte)


Session 3: Considering Workloads
Session Chairs: André Brinkmann (Johannes Gutenberg University Mainz)

Context-aware Prefetching for Near-Storage Accelerators
[Paper] [Slides]
Jian Zhang (Rutgers University), Marie Nguyen (Samsung), Sanidhya Kashyap (EPFL), Sudarsun Kannan (Rutgers University)

Advocating for Key-Value Stores with Workload Pattern Aware Dynamic Compaction
[Paper] [Slides]
Heejin Yoon, Jin Yang, Juyoung Bang (UNIST), Sam H. Noh (Virginia Tech), Young-ri Choi (UNIST)


Session 4: Keeping Data Correct
Session Chairs: Gala Yadgar (Technion Haifa)

Secure Archival is Hard... Really Hard
[Paper] [Slides]
Christopher Smith, Maliha Tabassum, Soumya Chowdary Daruru, Gaurav Kulhare, Arvin Wang (Stony Brook University), Ethan L. Miller (Pure Storage / University of California, Santa Cruz), Erez Zadok (Stony Brook University)

Asymmetric RAID: Rethinking RAID for SSD Heterogeneity
[Paper] [Slides]
Ziyang Jiao, Bryan S. Kim (Syracuse University)

Shadow Filesystems: Recovering from Filesystem Runtime Errors via Robust Alternative Execution
[Paper] [Slides]
Jing Liu, Xiangpeng Hao, Andrea Arpaci-Dusseau, Remzi Arpaci-Dusseau, Tej Chajed (UW-Madison)


Session 5: Minimizing Latencies
Session Chairs: Jason Xue (Mohamed bin Zayed University of Artificial Intelligence)

Breaking Barriers: Expanding GPU Memory with Sub-Two Digit Nanosecond Latency CXL Controller
[Paper] [Slides]
Donghyun Gouk (Panmnesia), Seungkwan Kang, Hanyeoreum Bae (KAIST), Eojin Ryu, Sangwon Lee, Dongpyung Kim (Panmnesia), Junhyeok Jang (KAIST), Myoungsoo Jung (Panmnesia and KAIST)

Improving Virtualized I/O Performance by Expanding the Polled I/O Path of Linux
[Paper] [Slides]
Dongjoo Seo (University of California, Irvine), Yongsoo Joo (Kookmin University), Nikil Dutt (University of California, Irvine)


Session 6: Storage and Machine Learning
Session Chairs: Zhichao Cao (ASU)

Rethinking Erasure-Coding Libraries in the Age of Optimized Machine Learning
[Paper] [Slides]
Jiyu Hu, Jack Kosaian, K. V. Rashmi (Carnegie Mellon University)

A Selective Preprocessing Offloading Framework for Reducing Data Traffic in DL Training
[Paper] [Slides]
Meng Wang, Gus Waldspurger (University of Chicago), Swaminathan Sundararaman (IBM Research)

Quantitative Analysis of Storage Requirement for Autonomous Vehicles
[Paper] [Slides]
Yuxin Wang, Yuankai He (University of Delaware), Ruijun Wang (Ford Motor Company), Weisong Shi (University of Delaware)


Session 7: Storage Technologies
Session Chairs: Janki Bhimani (FIU)

Can ZNS SSDs be Better Storage Devices for Persistent Cache?
[Paper]
Chongzhuo Yang, Zhang Cao, Chang Guo, Ming Zhao, Zhichao Cao (Arizona State University)

Can Storage Devices be Power Adaptive?
[Paper] [Slides]
Dedong Xie, Theano Stavrinos, Kan Zhu, Simon Peter, Baris Kasikci, Thomas E. Anderson (University of Washington)

Life-after-Death: Exploring Thermal Annealing Conditions to Enhance 3D NAND SSD Endurance
[Paper]
Matchima Buddhanoy, Sudeep Pasricha, Biswajit Ray (Colorado State University)