Understanding Inconsistency in Azure Cosmos DB with TLA+

-

This paper, by Finn Hackett, Joshua Rowe, Markus Kuppe, appeared in International Conference on Software Engineering 2023. It presents a specification of Azure Cosmos DB consistency behavior as exposed to the clients. 

During my sabbatical at CosmosDB in 2018, I was involved in a specification of CosmosDB as exposed to the clients.  The nice thing about these specs is that they didn't need to model internal implementation but just captured the consistency semantics for clients precisely, rather than ambiguously like English explanations do. They aimed to answer the question of what kind of behavior should a client be able to witness while interacting with the service? The feedback at that time was that customers found this very useful.

This 2023 paper improves on our preliminary specs from 2018. It has this great opening paragraph, which echoes the experience of everyone that has painstakingly specified a distributed/concurrent system behavior.

Consistency guarantees for distributed databases are notoriously hard to understand. Not only can distributed systems inherently behave in unexpected and counter-intuitive ways due to internal concurrency and failures, but they can also lull their users into a false sense of functional correctness: most of the time, users of a distributed database will witness a much simpler and more consistent set of behaviors than what is actually possible. Only timeouts, fail-overs, or other rare events will expose the true set of behaviors a user might witness. Testing for these scenarios is difficult at best: reproducing them reliably requires controlling complex concurrency factors, latency variations, and network behaviors. Even just producing usable documentation for developers is fundamentally challenging and explaining these subtle consistency issues via documentation comes as an additional burden to distributed system developers and technical writers alike.

I will focus on the TLA+ models developed by this work, as they are very instructive for intermediate/advanced TLA+ users. Also, the specifications provided in the paper generalizes well, so it is a good place to start with when modeling datastores. 


CosmosDB.tla

This spec models the multi-region Cosmos DB as a single log, which regulate the replication and failure behavior on this abstract log. I first took issue with this single log modeling. I thought this would be too simple, and would not be able to capture the behavior of a multi-regional deployment, with logs in each region. But when I challenged myself to find a client-exposed behavior of the multiple log system that would not be captured by the single log spec here, I was at a loss. Such is the power of nondeterminism. The single log, when instantiated nondeterministically is able to capture all those behaviors. For example, if the client in one region reads with eventual consistency, it is possible that the read value will later disappear if it wasn't part of the globally-committed log. Well, the single log spec allows truncation of non-committed suffix on the log, and captures this behavior.

CosmosDB.tla leverages the IncreaseReadIndexAndOrCommitIndex and TruncateLog actions to model this log. The commitIndex is the point in the log where all previous entries are replicated at a global majority of replicas, and are therefore durable due to consensus. The readIndex trails the commitIndex in the log potentially arbitrarily, and indicates the value is available at every region. Occasionally the spec will increase the readIndex and commitIndex of the log. TruncateLog can delete the suffix of the log upto reaching the commitIndex.

Note that the write action and read action are not exercised from this spec; those come in the `CosmosDBProps.tla` and `CosmosDBLinearizability.tla` specs that instantiate this spec, and use it to check consistency.


CosmosLinearizability.tla

This spec provides a brief linearizability test for the CosmosDB spec. In the first part of the file, it defines LinSpec which instantiates CosmosDB.tla spec and invokes CosmosDB!Init and CosmosDB!Next actions to exercise it. LinNext also introduces the write action, finally, to be exercised on CosmosDB spec.

The second part of the file imposes DictSpec as a property for LinSpec to satisfy. DictSpec lists all linearizable behavior of a simple dictionary writing values to keys and exposing results. By giving DictSpec in the Properties section of CosmosLinearizability.cfg, we are claiming that LinSpec refines DictSpec.

The interesting thing here is the dictView. We defined it almost like a materialized view in LinSpec figure. And here in this figure, we seed that DictSpec maintains another dictView expression. DictView is not even a variable, but it gets updated through the DictSpec. If the two dictViews contradict, this is flagged as property violation by the TLA model checker, and the trace is given as a counterexample.


CosmosProps.tla

This spec exercises write and read actions on CosmosDB.tla which it extends, and checks strong-consistency (aka linearizability) and causal-consistency properties on the MDB.tla spec.  

The first part of the spec extends the spec in the CosmosDB.tla with write action. The second part is about invariants and properties this spec satisfies for strong-consistency, bounded-staleness, session-consistency, and consistent-prefix/eventual consistency. Session-consistency read/writes is accomplished through session-tokens that are written to writeHistory variable in this spec. The properties checked are pretty general.  


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 >>