SOSP21 conference (Day 1)

-

SOSP, along with OSDI, is the premiere conference in computer systems. SOSP was held biannually; I had attended SOSP19 in person, and shared notes and paper summaries from that. 

This year SOSP is virtual, which made it a lot easier to travel to. It is nice attending the conference from the convenience of your home. The experience is not too inferior to physical conference attending if you take the convenience factor into account. 

Here are some notes from SOSP21. If you find and interesting paper, you can dig in, because:


Opening

The conference opened with announcements from program committee (PC) chairs. SOSP21 had 348 submissions from 2078 authors, resulting on avg 6 authors per paper. 54 out of 348 papers are accepted. Reviews were conducted by 64 PC members, who produced 1500+ reviews.

The trend is clear, the number of submissions are growing steeply.  Last year OSDI had announced it is going annual.  In the second day of SOSP it is announced that SOSP is going annual, starting from 2023.

To provide consistency across PCs, the OSDI rejected paper reviews were submitted to the conference along with those papers. The SOSP reviewers first reviewed the papers themselves, then saw OSDI reviews. This provided more consistency/continuation across submissions. The feedback is to make this optional rather than mandatory in the future.

Another experiment was to allow appendices in submissions. The appendices will be included in the cameraready, with the appendix marked as not peer-reviewed. This will help those who want to dive deeper into the paper. 

Best paper awards went to:

SOSP had put serious effort on artifact evaluation, which involved 91 reviewers, 171 reviews, 4.3 reviews per artifact. This resulted in issuing of  38 artifacts evaluated certificates and 27 results reproduced certificates.

The distinguished artifact award went to Rudra: finding memory safety bugs in Rust at the ecosystem scale. You can find results of the artifact evaluation, including links to all of the artifacts and summaries of the work done by the AEC at: https://sysartifacts.github.io/sosp2021/results.html


Logistics

Zoom was used for the conference. Audio/video is set to be off for audience by default. I didn't know this until the announcement was made. I was confused as I tried searching in vain to find the audio controls to mute myself. But I agree that this is a good setup for avoiding accidental zoom-bombing. When someone needs to ask a question, they are promoted to get access to video and audio. 

The poster session was done through gathertown. This was awesome. Gathertown overlays a physical space, where you run into people, look at posters. This felt like going to the conference, running people at the hallways. I enjoyed it. 


BFT session

Basil Breaking up BFT with ACID transactions

BIDL: A High-throughput, Low-latency Permissioned Blockchain Framework for Datacenter Networks

Kauri: Scalable BFT Consensus with Pipelined Tree-Based Dissemination and Aggregation


Bug finding session

iGUARD: In-GPU Advanced Race Detection

The title may be misleading. This is not a GPU application for computer vision-based race detection from photos/videos. GPU applications are growing. Every gpu generation is growing in size and complexity. GPUs introduce advance synchronization for programming. Sync only within a subset of threads, warp (32 thread), threadblock (1000 threads), or grid (even more threads). Independent thread scheduling (ITS). Races in CUDA code due to insufficient scopes. 12 races caught in popular. Anyways, it was great to see race issues being discussed in SOSP. 


Snowboard: Finding Kernel Concurrency Bugs through Systematic Inter-thread Communication Analysis

Finding concurrent inputs is challenging. Snowboard a tool for finding kernel concurrency bugs. It analyzes potential memory communications (pmc). It finds pmcs, prioritizes pmcs, and test them to find bugs.


Rudra: Finding Memory Safety Bugs in Rust at the Ecosystem Scale (Distinguished Artifact Award)

This work identified 3 bug patterns in "unsafe" Rust: panic safety bug, higher order invariant bug, and send/sync variance bugs.


Consistency

Witcher: Systematic Crash Consistency Testing for Non-Volatile Memory Key-Value Stores

NVM is here to stay. Two main operations: flush operation writes back a cache line to memory, fence operation provides ordering guarantee between flushes. They did an NVM software survey, and categorized things as persistence ordering bugs, persistence atomicity bugs, and persistence performance bugs. They use  rules for inferring likely correctness conditions specific to NVM programming. They then use output equivalence checking to validate a crashed state (this is not NVM specific). The tool takes program and test-case as input, do an  llvm compiler pass, and  run instrumented bytecode.


Understanding and Detecting Software Upgrade Failures in Distributed Systems

There is a tradeoff between safe versus fast upgrade. Developers use canary deployment to achieve some safety, but 20% of those take add 1000+ minutes to deployment. They did a study focusing on upgrade failures: 2/3 of upgrage failures were only caugth after software release. In 2/3rd of upgrade failures, the cause was data format and data semantic incompatibilities. 20% of syntax incompatibilities are defined by serialization libraries or enum.


Crash Consistent Non-Volatile Memory Express


Databases

Cuckoo Trie: Exploiting Memory-Level Parallelism for Efficient DRAM Indexing

Cuckoo trie: an ordered index that exploits memory-level parallelism (mlp). Start with a trie, store nodes in a hash table, no pointers are needed, you can read nodes in parallel. https://youtu.be/_YmXL2gYzRI


Regular Sequential Serializability and Regular Sequential Consistency 

Strict serializability guarantees total order of transactions: The order respects causality and the order also respects some real-time: reads must return up-to-date values. They introduce regular sequential serializability (rss). Informally any application inv that holds with a strictly serializable service also holds with an rss service. https://youtu.be/GauK7QhaAEQ


Caracal: Contention Management with Deterministic Concurrency Control 

Comments

Post a Comment

Popular posts from this blog

Learning about distributed systems: where to start?

-
This is definitely not a "learn distributed systems in 21 days" post. I recommend a principled, from the foundations-up, studying of distributed systems, which will take a good three months in the first pass, and many more months to build competence after that. If you are practical and coding oriented you may not like my advice much. You may object saying, "Shouldn't I learn distributed systems with coding and hands on? Why can I not get started by deploying a Hadoop cluster, or studying the Raft code." I think that is the wrong way to go about learning distributed systems, because seeing similar code and programming language constructs will make you think this is familiar territory, and will give you a false sense of security. But, nothing can be further from the truth. Distributed systems need radically different software than centralized systems do.  --A. Tannenbaum This quotation is literally the first sentence in my distributed systems syllabus. Inst
Read more >>

Hints for Distributed Systems Design

-
This is with apologies to Butler Lampson, who published the " Hints for computer system design " paper 40 years ago in SOSP'83. I don't claim to match that work of course. I just thought I could draft this post to organize my thinking about designing distributed systems and get feedback from others. I start with the same  disclaimer Lampson gave. These hints are not novel, not foolproof recipes, not laws of design, not precisely formulated, and not always appropriate. They are just hints.  They are context dependent, and some of them may be controversial. That being said, I have seen these hints successfully applied in distributed systems design throughout my 25 years in the field, starting from the theory of distributed systems (98-01), immersing into the practice of wireless sensor networks (01-11), and working on cloud computing systems both in the academia and industry ever since. These heuristic principles have been applied knowingly or unknowingly and has proven
Read more >>

Foundational distributed systems papers

-
I talked about the importance of reading foundational papers last week. To followup, here is my compilation of foundational papers in the distributed systems area. (I focused on the core distributed systems area, and did not cover networking, security, distributed ledgers, verification work etc. I even left out distributed transactions, I hope to cover them at a later date.)  I classified the papers by subject, and listed them in chronological order. I also listed expository papers and blog posts at the end of each section. Time and State in Distributed Systems Time, Clocks, and the Ordering of Events in a Distributed System. Leslie Lamport, Commn. of the ACM,  1978. Distributed Snapshots: Determining Global States of a Distributed System. K. Mani Chandy Leslie Lamport, ACM Transactions on Computer Systems, 1985. Virtual Time and Global States of Distributed Systems.  Mattern, F. 1988. Practical uses of synchronized clocks in distributed systems. B. Liskov, 1991. Expository papers
Read more >>

Metastable failures in the wild

-
Image
This paper appeared in OSDI'22. There is a great summary of the paper by Aleksey (one of the authors and my former PhD student, go Aleksey!). There is also a great conference presentation video from Lexiang. Below I will provide a brief overview of the paper followed by my discussion points. This topic is very interesting and important, so I hope you have fun learning about this. Metastability concept and categories Metastable failure is defined as permanent overload with low throughput even after the fault-trigger is removed. It is an emergent behavior of a system, and it naturally arises from the optimizations for the common case that lead to sustained work amplification. In this paper, the authors are able to capture/abstract the system behavior of interest in terms of two parameters, the load and capacity. If the load is above capacity, you have work piling up, right? Or if the capacity drops under the sustained load level, the same effect, right? Both of these create  a tem
Read more >>

The demise of coding is greatly exaggerated

-
Image
NVDIA CEO Jensen Huang recently made very contraversial remarks : "Over the course of the last 10 years, 15 years, almost everybody who sits on a stage like this would tell you that it is vital that your children learn computer science, and everybody should learn how to program. And in fact, it’s almost exactly the opposite. It is our job to create computing technology such that nobody has to program and that the programming language is human. Everybody in the world is now a programmer.  This is the miracle of artificial intelligence." I am not going to wise crack and say that this is power poisioning and this is what happens when your company valuation more than triples in a year and surpasses Amazon and Google. (Although I don't discount this effect completely.) Jensen is very smart and also has some great wisdom , so I think we should give this the benefit of doubt and try to respond in a thoughtful manner.  A response is warranted because this statement got a lot of p
Read more >>

The end of a myth: Distributed transactions can scale

-
Image
This paper appeared in VLDB'17. The paper presents NAM-DB, a scalable distributed database system that uses RDMA (mostly 1-way RDMA) and a novel timestamp oracle to support snapshot isolation (SI) transactions. NAM stands for network-attached-memory architecture, which leverages RDMA to enable compute nodes talk directly to a pool of memory nodes. Remote direct memory access (RDMA) allows bypassing the CPU when transferring data from one machine to another. This helps relieve a major factor in scalability of distributed transactions: the CPU overhead of the TCP/IP stack. With so many messages to process, CPU may spend most of the time serializing/deserializing network messages, leaving little room for the actual work. We had seen this phenomena first hand when we were researching the performance bottlenecks of Paxos protocols. This paper reminds me of the "Is Scalable OLTP in the Cloud a Solved Problem? (CIDR 2023)" which we reviewed recently. The two papers share one
Read more >>

Scalable OLTP in the Cloud: What’s the BIG DEAL?

-
Image
This paper is from Pat Helland, the apostate philosopher of database systems, overall a superb person, and a good friend of mine. The paper appeared this week at CIDR'24. (Check out the program for other interesting papers). The motivating question behind this work is: " What are the asymptotic limits to scale for cloud OLTP (OnLine Transaction Processing) systems? " Pat says that the CIDR 2023 paper "Is Scalable OLTP in the Cloud a Solved Problem?" prompted this question.  The answer to the question? Pat says that the answer lies in the joint responsibility of database and the application. If you know of Pat's work, which I have summarized several in this blog , you would know that Pat has been advocating along these lines before. But this paper provides a very crisp, specific, concrete answer. Read on for my summary of the paper. Disclaimer: This is a wisdom and technical information/detail packed 13-page paper, so I will try my best to summarize the sa
Read more >>

SIGMOD panel: Future of Database System Architectures

-
Image
I mentioned this panel in my SIGMOD/PODS day 2 writeup.  The panel consisted of (from right to left) Gustavo Alonso (ETH), Swami Sivasubramanian (AWS), Anastasia Ailamaki (EPFL), Raghu Ramakrishnan (Microsoft), and Sam Madden (MIT). Sailesh Krishnamurthy from Google was unable to attend, so Anastasia who is on sabbatical at Google, also covered for him. The panel started with 5 minutes speeches from participants. The panel was lively and playful. The panelists tried to make their speeches controversial, and have zingers for each other.   Sam talked about the demise of DBMS monolith. He said, the Oracle of 90s is dead because it was incompatible with modern data ecosystem, and the modern DBMSs are shifting towards disaggregated designs. When Sam was a PhD student under Eric Brewer, Eric said we should build disaggregated databases, and people said that won't happen. But we are there now: Disaggregation has arrived! While Sam mentioned about the benefits of disaggregation, he also ca
Read more >>

Why I blog

-
Image
My blog has been going for 14 years now, and has just passed 4 million pageviews. Yay! I remember the 1 million pageviews moment in 2017 ! The main reason I was able to persist for so long is because I blog for selfish reasons. Let me try to unpack why I blog, and why I keep blogging. I write for myself The audience I have in mind is myself. I blog to clarify my understanding and thinking about a topic.  Reading a research/technical paper is already time consuming. I can't do it in less than 4 hours. Period. I love learning. And I am fortunate that I get to read research papers as part of my work. I double-dip on this effort to blog about them, to improve my understanding of these papers. Writing a blog post is the final step in my pipeline for reading a paper. I think my blog reviews of papers hits a good niche. Research papers are written for the wrong audience (or rather maybe the right audience but for the wrong reason): they are written to please 3 specific expert reviewers
Read more >>

There is plenty of room at the bottom

-
Image
This is a pun on the saying "there is always room at the top". This is also the title of a famous Feynman lecture from 1959 , where he made a case for nanotechnology.  In this post, I will try to argue that there is plenty of room at the bottom for distributed algorithms. Most work on distributed algorithms are done at a high-level abstraction plane. These high level solutions do not transfer well to the implementation level, and this opens a lot of space to explore at the implementation level. But this is not just an opportunistic argument. It is imperative to target the implementation level with our distributed algorithms work, otherwise they remain as theoretical, unused, inapplicable.  Let me try to demonstrate using consensus protocols as examples. Someone else can make the same case using another subdomain. Be mindful of what is swept under the rug What is succinct at the high level could be very hard to implement and get right at the low level. Paxos seems simple , bu
Read more >>