Best of Metadata in 2022

-

It is time to reflect back on the year. Here are some selected papers I covered in 2022.

Production systems

Amazon Aurora: Design Considerations + On Avoiding Distributed Consensus for I/Os, Commits, and Membership Changes
Aurora provides a a high-throughput relational database on AWS cloud by decoupling the storage to a multi-tenant scale-out storage service. Each database instance acts as a SQL endpoint and supports query processing and transactions. Many database functions, such as redo logging, materialization of data blocks, garbage collection, and backup/restore, are offloaded to the storage nodes. Aurora performs replication among the storage nodes by pushing the redo log; this reduces networking traffic and enables fault-tolerant storage that heals without database involvement.


Amazon DynamoDB: A Scalable, Predictably Performant, and Fully Managed NoSQL Database Service (USENIX ATC 2022)
DynamoDB has massive scale; it powers Alexa, Amazon.com sites, and all Amazon fulfillment centers. Hundreds of thousands of customer applications also use DynamoDB. DynamoDB's architecture does not share much with that of the Dynamo system (2007). DynamoDB uses MultiPaxos for replication. This paper does not focus on the DynamoDB architecture anyways; it is more about achieving operational excellence while operating arguably the largest scale cloud database ever. DynamoDB needs to deal with operational issues related to fairness, traffic imbalance across partitions, monitoring, and automated system operations without impacting availability or performance. Reliability and predictability of performace is essential, as even the slightest disruption can significantly impact customers.


RAMP-TAO: Layering Atomic Transactions on Facebook’s Online TAO Data Store (VLDB'21)
This paper discusses the TAO graph datastore system used by Facebook and the motivation for adding a read transaction API to it. It also covers the design, implementation, and evaluation of the RAMP-TAO protocol, including the determination of the most appropriate isolation level for it.


SQLite: Past, Present, and Future
SQLite is the most widely deployed database engine with more than one trillion SQLite databases in active use. SQLite is a single node and (mostly) single threaded online transaction processing (OLTP) database. It has an in-process/embbedded design, and a standalone (no dependencies) codebase ...a single C library consisting of 150K lines of code. Being designed as an OLTP database, SQLite employs row-oriented execution and a B-tree storage format. However, there is a growing need for efficient in-process online analytical processing (OLAP). This paper delves into analytical data processing on SQLite, identifying key bottlenecks and implementing suitable solutions. As a result of the optimizations implemented, SQLite is now up to 4.2X faster on the Star Schema Benchmark (SSB). This is a sweet little paper (befitting SQLite's fame).


CockroachDB: The Resilient Geo-Distributed SQL Database
CockroachDB consists of a distributed SQL layer on top of a distributed key-value store. The key value store is laid out in a single monolithic ordered key space divided it into ranges. The ranges are the unit of replication. CockroachDB deploys a Raft group per each range for replication. CockroachDB uses multi-version concurrency control. This paper explains CockroachDB's 2PC+2PL based distributed transactions protocol and its serializable multiversion concurrency control protocol.


FoundationDB: A Distributed Unbundled Transactional Key Value Store (Sigmod 2021)
FoundationDB is a transactional key-value store that supports multi-key strictly serializable transactions across its entire key-space. It is used at Apple and Snowflake, due to its consistency, robustness and availability for storing user data, system metadata and configuration, and other critical information. The main idea in FDB is to decouple transaction processing from logging and storage. Such an unbundled architecture enables the separation and horizontal scaling of both read and write handling. The transaction system combines optimistic concurrency control and multi-version concurrency control and achieves strict serializability or snapshot isolation if desired.


Database Research

Decoupled Transactions: Low Tail Latency Online Transactions Atop Jittery Servers (CIDR 2022)
This is a Pat Helland paper. Pat's papers are always remarkable and full of wisdom. The paper proposes a novel design to dampen application visible jitter in transactional databases, and to scale OLTP SQL databases to tens of servers with predictable response time while running in a largely unpredictable environment. The design leverages quorums to solve the jitter problem. It is a very technical and dense paper, despite the colloquial writing style. Appendix D, Sections 15.3 and 15.4 show how to address the jitter free liveness check via log and the jitter free concurrent log fencing. I had gotten these wrong about the paper, and criticized the paper with my limited understanding of the design. It took many hours of discussion with Pat to get a better understanding of the protocols in the paper. I owe Pat a revised write-up but I have been procrastinating on this. Well, I think 2023 will be the year. This paper deserves very careful studying.  


Anna: A Key-Value Store For Any Scale
This paper introduces Anna, a CALM/CRDT implementation of a distributed key-value system both at the data structure level as well as system architecture and transaction protocol levels. Conventional wisdom says that software designed for one scale point needs to be rewritten when scaling up by 10x. Anna aims to disprove this by showing how a key-value storage system can be architected to scale across many orders of magnitude. Anna is a partitioned, multi-mastered key-value system that achieves high performance and elasticity via wait-free execution and coordination-free consistency. Anna employs coordination-free actors that perform state update via merge of lattice-based composite data structures. Anna can provide consistency up to causal consistency. It cannot provide strong consistency at key level, and nothing stronger than read-committed at the multi-key level.


Calvin: Fast Distributed Transactions for Partitioned Database Systems
The name Calvin alludes to the predestination doctrine in Calvinism. Calvin is all about pre-ordering transactions in logs in batches, and then deterministically executing this log in the same order in replicas. The key technical feature that allows Calvin scalability in the face of distributed transactions is a deterministic ordering/locking mechanism that enables the elimination of distributed commit protocols. Since the schedulers get a good look of which transactions are coming next in this log, they can perform optimizations by pulling some independent transactions locally and executing them in parallel in worker partitions with no ill effect. Calvin achieves good performance thanks to the batching and optimizations pulled by the schedulers by looking ahead to transactions in the log. The downsides are the loss of conversational transaction execution as in SQL and increased latency due to batch processing. FaunaDB implements Calvin, and there has been several followup work to Calvin. It is an important technique to learn.


Fast Serializable Multi-Version Concurrency Control for Main-Memory Database Systems
The paper describes the addition of Multi-Version Concurrency Control (MVCC) to the Hyper database, and discusses at length how they implement serializability efficiently with little overhead compared to snapshot isolation. The paper says that Hyper is like Hekaton a closely related in-memory OLTP database in that, in order to preserve scan performance, new transactions are not allocated a separate place. The most uptodate version of the database is kept all in the same place. But, you can reach older versions of a record (that are still in use by active transactions) by working backwards from the record using the delta versions maintained. The biggest improvement in Hyper over Hekaton is the more efficient and tidier implementation of multiversion concurrency control, especially for providing serializability with little overhead.


Timestamp-based Algorithms for Concurrency Control in Distributed Database Systems
This paper is from 1980, yet it talked about distributed databases, and laid out a forward looking distributed database management system architecture, which is very relevant and in use even today. I really loved the forward looking vision and clear thinking and presentation in the paper. The paper presents several timestamp order based optimistic concurrency control protocols. As Doug Terry talks about in his Fast19 keynote, DynamoDB transactions employ this timestamp-ordered transactions idea.


Seeing is Believing: A Client-Centric Specification of Database Isolation
The paper provides a dense but rewarding read. It presents a new way to reason about transaction-isolation based on application-observable states, in lieu of prior history-based approaches. While history-based approaches prescribed operational reasoning on transaction isolation semantics, the proposed state-based approach opened the way for invariant-based reasoning for this. This approach freed isolation definitions from implementation-specific details (timestamps, replicas) and opens the way to find alternate implementations. A followup work provided TLA+ formalization of the state-based isolation model introduced here.


Misc

Efficient Replication via Timestamp Stability (EuroSys 2021)
This paper introduces Tempo, a simpler more streamlined leaderless Paxos variant for implementing a state machine replication protocol. Recently we have been seeing a convergence of the leaderless SMR ideas with loosely synchronized clocks. Tempo brings us closer to reaping the benefits of synchronized system model, while still being safe to the face of bouts of full asynchrony in the system. Accord protocol improves on Tempo and is implemented in Cassandra to provide distributed/leaderless transactions.


TLA+ Conference and StrangeLoop 2022
This is a writeup of the TLA+ conference held as part of Strange Loop. There were several awesome talks on recent developments in the TLA+ space.


Checking statistical properties of protocols using TLA+
The TLA+ model checker has been extended with a new mode (called "generate") to enable obtaining statistical properties from the models. Checking statistics is very useful, because ensuring correctness addresses only one part of our concerns with practical systems.


Previous years in review

Best of metadata in 2021


Best of metadata in 2020


Best of metadata in 2019


Best of metadata in 2018


Research, writing, and career advice

Comments

Post a Comment

Popular posts from this blog

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 >>

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 >>

Making database systems usable

-
Image
C. J. Date's Sigmod 1983 keynote, "Database Usability", was prescient. Usability is the most important thing to the customers. They care less about impressive benchmarks or clever algorithms, and more about whether they can operate and use a database efficiently to query, update, analyze, and persist their data with minimal headache. (BTW, does anyone have a link to the contents of this Sigmod'83 talk? There is no transcript around, except for this short abstract .) The paper we cover today is from Sigmod 2007. It takes on the database usability problem raised in that 1983 keynote head-on, and calls out that the king is still naked.  Let's give some context for the year 2007. Yes, XML format was still popular then. The use-case in the paper is XQuery. The paper does not contain any  reference to json. MongoDB would be released in 2009 with the document model; and that seems to be great timing for some of the usability pains mentioned in the paper! Web 2.0 was in
Read more >>

Looming Liability Machines (LLMs)

-
As part of our zoom reading group ( wow, 4.5 years old now ), we discussed a paper that uses LLMs for automatic root cause analysis (RCA) for cloud incidents. This was a pretty straightforward application of LLMs. The proposed system employs an LLM to match incoming incidents to incident handlers based on their alert types, predicts the incident's root cause category, and provides an explanatory narrative. The only customization is through prompt-engineering. Since this is a custom domain, I think a more principled and custom-designed  machine learning system would be more appropriate rather than adopting LLMs. Anyways, the use of LLMs for RCAs spooked me vicerally. I couldn't find the exact words during the paper discussion, but I can articulate this better now. Let me explain. RCA is serious business Root cause analysis (RCA) is the process of identifying the underlying causes of a problem/incident, rather than just addressing its symptoms. One RCA heuristic is asking 5 Why&#
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 >>

Advice to the young

-
Image
I notice I haven't written any advice posts recently. Here is a collection of my advice posts pre 2020. I've been feeling all this elderly wisdom pent up in me, ready to pour at any moment. So here it goes. Get ready to quench your thirst from my fount of wisdom. No man, think for yourself, only get what works for you. It is called foundations, not theory Foundations of computer science (or rather any field of study) are the most important topics you can learn. These lay down the frame of thinking/perspective for that area of study. Yet, I am saddened to hear these called as "theory", and labeled as "unpractical". This couldn't be farther from the truth. Take a look at how I recommend studying distributed systems . Don't you dare call this "theory" and "unpractical". This lays the bedrock that you build your practice on. Don't skimp on the foundations. Don't build your home on quicksand. Keep your hands dirty, your mind cl
Read more >>

Linearizability: A Correctness Condition for Concurrent Objects

-
Image
This paper is from Herlihy and Wing appeared in ACM Transactions on Programming Languages and Systems 1990. This is the canonical reference for the linearizability definition. I had not read this paper in detail before, so I thought it would be good to go to the source to see if there are additional delightful surprises in the original text. Hence, this post. I will dive into a technical analysis of the paper first, and then discuss some of my takes toward the end. I had written an accessible explanation of linearizability earlier; you may want to read that first. I will assume an understanding of linearizability to keep this review at reasonable length. Introduction I love how the old papers just barge in with the model, without bothered by pleasantries such as motivation of the problem. These are the first two sentences of the introduction. "A concurrent system consists of a collection of sequential processes that communicate through shared typed objects . This model encompass
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 >>

Understanding the Performance Implications of Storage-Disaggregated Databases

-
Image
Storage-compute disaggregation in databases has emerged as a pivotal architecture in cloud environments, as evidenced by Amazon ( Aurora ), Microsoft ( Socrates ), Google (AlloyDB), Alibaba ( PolarDB ), and Huawei (Taurus). This approach decouples compute from storage, allowing for independent and elastic scaling of compute and storage resources. It provides fault-tolerance at the storage level. You can then share the storage for other services, such as adding read-only replicas for the databases. You can even use the storage level for easier sharding of your database. Finally, you can also use this for exporting a changelog asynchronously to feed into peripheral cloud services, such as analytics. Disaggregated architecture was the topic of Sigmod 23 panel . I think this quote summarizes the industry's thinking on the topic. "Disaggregated architecture is here, and is not going anywhere. In a disaggregated architecture, storage is fungible, and computing scales independently.
Read more >>

Designing Data Intensive Applications (DDIA) Book

-
Image
We started reading this book as part of Alex Petrov's book club . We just got started, so you can join us, by joining the discord channel above. We meet Wednesday's 11am Eastern Time.  Previously we had read transaction processing book by Grey and Reuters. This page links to my summaries of that book. Chp 1. Reliable, Scalable, and Maintainable Applications I love the diagrams opening each chapter. Beautiful! The first chapter consists of warm up stuff. It talks about the definitions of reliabilty, scalability, and maintainability. It is still engaging, because  it is written with an educator and technical blogger voice, rather than a dry academic voice. This book came out on 2017. Martin is working on the new version. So if you have comments for things to focus on for the new version, it would be helpful to collect them in a document and email it to Martin. For example, I am curious about how the below paragraph from the Preface will get revised with 8 more years of hindsight:
Read more >>