The schedule is also in our Google Calendar.

• Omni-Paxos: Breaking the Barriers of Partial Connectivity (EuroSys’23)
  • Authors: Harald Ng, Seif Haridi, Paris Carbone
  • What: Multi-Paxos flavor with better tolerance for partial network partitions
  • When: August 30th
• Hyrax: Fail-in-Place Server Operation in Cloud Platforms (OSDI’23)
  • Authors: Jialun Lyu, Marisa You, Celine Irvene, Mark Jung, Tyler Narmore, Jacob Shapiro, Luke Marshall, Savyasachi Samal, Ioannis Manousakis, Lisa Hsu, Preetha Subbarayalu, Ashish Raniwala, Brijesh Warrier, Ricardo Bianchini, Bianca Schroeder, Daniel S. Berger
  • What: “Datacenter stack that enables compute servers with failed components to continue hosting VMs while hiding the underlying degraded capacity and performance”
  • When: September 6th
• Aggregate VM: Why Reduce or Evict VM’s Resources When You Can Borrow Them From Other Nodes? (EuroSys’23)
  • Authors: Ho-Ren Chuang, Karim Manaouil, Tong Xing, Antonio Barbalace, Pierre Olivier, Balvansh Heerekar, Binoy Ravindran
  • What: VMs with resources aggregated across multiple physical hosts
  • When: September 13th
• Take Out the TraChe: Maximizing (Tra)nsactional Ca(che) Hit Rate (OSDI’23)
  • Authors: Audrey Cheng, David Chu, Terrance Li, Jason Chan, Natacha Crooks, Joseph M. Hellerstein, Ion Stoica
  • What: Object hit rate does not matter if transactions still touches non-cached objects.
  • When: September 20th
• Chardonnay: Fast and General Datacenter Transactions for On-Disk Databases (OSDI’23)
  • Authors: Tamer Eldeeb, Xincheng Xie, Philip A. Bernstein, Asaf Cidon, Junfeng Yang
  • What: Another fast 2PC paper
  • When: September 27th
• Distributed Transactions at Scale in Amazon DynamoDB (ATC’23)
  • Authors: Joseph Idziorek, Alex Keyes, Colin Lazier, Somu Perianayagam, Prithvi Ramanathan, James Christopher Sorenson III, Doug Terry, Akshat Vig
  • What: See the title
  • When: October 4th
• Palette Load Balancing: Locality Hints for Serverless Functions (EuroSys’23)
  • Authors: Mania Abdi, Sam Ginzburg, Charles Lin, Jose M Faleiro, Íñigo Goiri, Gohar Irfan Chaudhry,Ricardo Bianchini, Daniel S. Berger, Rodrigo Fonseca
  • What: Data locality hints for better caching in serverless runtimes.
  • When: October 18th
• Oakestra: A Lightweight Hierarchical Orchestration Framework for Edge Computing (ATC’23)
  • Authors: Giovanni Bartolomeo, Mehdi Yosofie, Simon Bäurle, Oliver Haluszczynski, Nitinder Mohan, Jörg Ott
  • What: Orchestration frameworks (e.g. Kubernetes) for the edge
  • When: October 25th
• On-demand Container Loading in AWS Lambda (ATC’23)
  • Authors: Marc Brooker, Mike Danilov, Chris Greenwood, Phil Piwonka
  • What: See the title
  • When: November 1st
• Model Checking Guided Testing for Distributed Systems (EuroSys’23)
  • Authors: Dong Wang, Wensheng Dou, Yu Gao, Chenao Wu, Jun Wei, Tao Huang
  • What: Testing guided by the state space generated by the model checker
  • When: November 8th

Reading Group

Our reading group takes place over Zoom every Wednesday at 2:00 pm EST. We have a slack group where we post papers, hold discussions, and most importantly manage Zoom invites to paper discussions. Please join the slack group to get involved!

A new set of papers for spring and early summer!

• Transactions Make Debugging Easy [CIDR’23]
  • Authors: Qian Li, Peter Kraft, Michael Cafarella, Çağatay Demiralp, Goetz Graefe, Christos Kozyrakis, Michael Stonebraker, Lalith Suresh, and Matei Zaharia
  • What: Everything is a database transaction, including debugging
  • When: April 12th
• Perseus: A Fail-Slow Detection Framework for Cloud Storage Systems [FAST’23]
  • Authors: Ruiming Lu, Erci Xu, Yiming Zhang, Xiamen University; Fengyi Zhu, Zhaosheng Zhu, Mengtian Wang, Zongpeng Zhu, Guangtao Xue, Jiwu Shu, Minglu Li, Jiesheng Wu
  • What: Detecting fail-slow failures (i.e., machine/node slowdowns) in storage systems
  • When: April 19th
• Developer’s Responsibility or Database’s Responsibility? Rethinking Concurrency Control in Databases [CIDR’23]
  • Authors: Chaoyi Cheng, Mingzhe Han, Nuo Xu, Spyros Blanas, Michael D. Bond, and Yang Wang
  • What: Study of 93 concurrency bugs in database applications
  • When: April 26th
• SelfTune: Tuning Cluster Managers [NSDI’23]
  • Authors: Ajaykrishna Karthikeyan, Nagarajan Natarajan, Gagan Somashekar, Lei Zhao, Ranjita Bhagwan, Rodrigo Fonseca, Tatiana Racheva, Yogesh Bansal
  • What: Automatically tuning cluster manager parameters/setting for ever-changing cluster state.
  • When: May 3rd
• FUSEE: A Fully Memory-Disaggregated Key-Value Store [FAST’23]
  • Authors: Jiacheng Shen, Pengfei Zuo, Xuchuan Luo, Tianyi Yang, Yuxin Su, Yangfan Zhou, and Michael R. Lyu
  • What: KV-store with memory-disaggregated storage and metadata
  • When: May 17th
• ROLEX: A Scalable RDMA-oriented Learned Key-Value Store for Disaggregated Memory Systems [FAST’23]
  • Authors: Pengfei Li, Yu Hua, Pengfei Zuo, Zhangyu Chen, and Jiajie Sheng
  • What: RDMA-based disaggregated memory store with learned index
  • When: May 24th
• Hydra: Serialization-Free Network Ordering for Strongly Consistent Distributed Applications [NSDI’23]
  • Authors: Inho Choi, Ellis Michael, Yunfan Li, Dan R. K. Ports, and Jialin Li
  • What: In-network ordering with distributed sequencers
  • When: May 31st
• Stabilizer: Geo-Replication with User-defined Consistency [ICDCS’22]
  • Authors: Pengze Li, Lichen Pan, Xinzhe Yang, Weijia Song, Zhen Xiao, Ken Birman
  • What: High-performance geo-replication library with user-defined consistency and separated control and data planes
  • When: June 7th
• Closing the B+-tree vs. LSM-tree Write Amplification Gap on Modern Storage Hardware with Built-in Transparent Compression [FAST’22]
  • Authors: Yifan Qiao, Xubin Chen, Ning Zheng, Jiangpeng Li, Yang Liu, Tong Zhang
  • What: Reducing B+-tress write amplification
  • When: June 14th
• Real-time Data Infrastructure at Uber
  • Authors: Yupeng Fu, Chinmay Soman
  • What: Uber’s data infrastructure and scalability challenges it faces
  • When: June 21st

With just four more papers to go in the DistSys Reading Group’s current batch, it is time to get the next set going. This round, we will have 10 papers that should last till the end of the spring semester. Our last batch was all about OSDI’20 papers, and this time around we will mix things around both in terms of the venues and paper recency. We will also start the batch with one foundational paper taken from Murat’s recent list of must-read classical papers in distributed systems. Without further ado, here is the list:

1. Distributed Snapshots: Determining Global States of a Distributed System – April 7th
2. Facebook’s Tectonic Filesystem: Efficiency from Exascale – April 14th
3. New Directions in Cloud Programming – April 21st
4. Paxos vs Raft: Have we reached consensus on distributed consensus? – April 28th
5. Protocol-Aware Recovery for Consensus-Based Storage – May 5th
6. chainifyDB: How to get rid of your Blockchain and use your DBMS instead – May 12th
7. XFT: practical fault tolerance beyond crashes – May 19
8. Cerebro: A Layered Data Platform for Scalable Deep Learning – May 26th
9. Multitenancy for Fast and Programmable Networks in the Cloud – June 2nd
10. Exploiting Symbolic Execution to Accelerate Deterministic Databases – June 9th

Reading Group

Our reading group takes place over Zoom every Wednesday at 2:00 pm EST. We have a slack group where we post papers, hold discussions, and most importantly manage Zoom invites to paper discussions. Please join the slack group to get involved!

This week we are on the 35th paper in our reading group. We will be discussing “Autoscaling Tiered Cloud Storage in Anna” from VLDB 2019. It is exciting that the reading group has managed to go through this many papers since we have started in April! 

Today I and Murat have sat down and picked the next few papers from the OSDI 2020 to discuss. Originally we planned to select just 10 OSDI papers for the next batch, but all the papers are so interesting, and it was so hard to narrow the list down. After some debating, we settled on 14 papers instead. Below is the list in the order the papers will appear in the reading group:

1. Fault-tolerant and transactional stateful serverless workflows – December 23rd
2. Aragog: Scalable Runtime Verification of Shardable Networked Systems – December 30th
3. Toward a Generic Fault Tolerance Technique for Partial Network Partitioning – January 6th
4. hXDP: Efficient Software Packet Processing on FPGA NICs – January 13th
5. Virtual Consensus in Delos – January 20th
6. A large scale analysis of hundreds of in-memory cache clusters at Twitter – January 27th
   
7. Cobra: Making Transactional Key-Value Stores Verifiably Serializable – February 3rd
   
8. Microsecond Consensus for Microsecond Applications  – February 10th
9. Performance-Optimal Read-Only Transactions – February 17th
10. Pegasus: Tolerating Skewed Workloads in Distributed Storage with In-Network Coherence Directories – February 24th
11. FlightTracker: Consistency across Read-Optimized Online Stores at Facebook – March 3rd
12. Heterogeneity-Aware Cluster Scheduling Policies for Deep Learning Workloads – March 10th
13. Sundial: Fault-tolerant Clock Synchronization for Datacenters – March 17th
14. Protean: VM Allocation Service at Scale – March 24th

This list will start on December 23rd, and will run us all the way to paper #50! Please join the reading group on slack for schedule, discussion, and Zoom links. The reading group Zoom is every Wednesday at 3:30 pm EST.