Martin Kleppmann
Researcher on distributed systems and local-first software at the University of Cambridge; wrote Designing Data-Intensive Applications.
Martin Kleppmann did not write this page. ppll compiled it from their own words, each linked to the source. Is this you? · Report a mistake
Latest Bluesky
What they believe
-
Reliability in a data system means fault tolerance above all: the system keeps working when a network link is cut or a node dies.
for me, reliability means fault tolerance primarily. So, meaning that a system should, on the whole, continue working even if like a network link is interrupted or a node crashes or something like that.
youtube.comSaid 22 Apr 2026This on korrents
The bold line is korrents’ wording; the quote is theirs. These one: what they wrote before what they said aloud, one per source, newest first.
Feed
As its own page →Hiding
4 November
2 November
28 October
27 October
18 October
14 October
13 October
12 October
7 October
28 September
27 September
22 April
From one piece Designing Data-intensive Applications with Martin Kleppmann 15 beliefs · youtube.com
-
Their words
In particular, for example, coverage of MapReduce was quite detailed in the first edition. But basically, MapReduce is dead. Nobody uses it anymore.
-
Their words
And so, the goal of this chapter is really just to give people the information in order to make an educated decision. But I don't want to make that decision for people. That's for businesses themselves to decide.
-
korrents.com
Clocks are right most of the time, and that is precisely why a distributed system may never rely on them.Their words
it's very easy to assume that your clocks are correct and most of the times the clocks are pretty correct but we just can't rely on it because actually they're just not precise enough uh on the whole.
+ 12 more
-
Their words
we just assume that there's no upper bound on how long it might take for a message to go over the network. So, you send a message, it might arrive within 100 microseconds, or it might take 10 years.
-
korrents.com
Replication earns its place even in a small system, because what it buys is fault tolerance rather than scale.Their words
And uh replication is still relevant even at smaller scales because that's for fault tolerance, that's not for scalability.
-
Their words
But, at least this sort of sharding across multiple machines is maybe become less of a pressing issue just because more and more workloads can just run on a single machine. Some people still have very large-scale workloads that do have to be sharded across multiple machines. So, it's not going away entirely.
-
Their words
Maybe as AI writes more and more code of our code, it's less about like the details of how you express logic in a particular programming language and much more about those kinds of high-level trade-offs.
-
Their words
So, what if geopolitics was to go horribly wrong and tensions escalate and Europe find itself suddenly locked out of US cloud services? I hope that doesn't happen. I still think it's fairly unlikely, but it's no longer unthinkable.
-
Their words
And so those are those places where I feel like knowing a bit about the the internals is actually like a superpower.
-
Their words
But, somebody still has to build those lower-level abstractions and from lower-level components. Somebody's got to implement the cloud services.
-
Their words
You're building higher-level business logic, actually I think it's just fine for people not to care about memory management.
-
Their words
And now suddenly people are building databases on top of object stores, for example. And now the replication happens at the object store level, no longer at the database level.
-
Their words
generally, like you just want the the cost and the computing capacity to be roughly proportional to the load that you have. And at the low end, that means actually being able to scale down to something that is extremely cheap to run.
-
korrents.com
Scalability is not a property a system has; it is the set of mechanisms it has for coping with a change in load.Their words
Scalability is just like what mechanisms we have for dealing with changes in load. If load increases, how can we add computing capacity to a system, for example, so that the system still continues working?
-
Their words
for me, reliability means fault tolerance primarily. So, meaning that a system should, on the whole, continue working even if like a network link is interrupted or a node crashes or something like that.
8 December 2025
5 July 2024
4 January 2024
12 October 2022
3 January 2022
-
Lovedaffiliate linkrcmnd.app
The Future of Fusion EnergyTheir words
The book provides an excellent, detailed answer to these questions, and more. It’s the best kind of popular science book: you don’t need a physics degree to read it, but it doesn’t fob you off with oversimplified hand-waving either; all of the core arguments are convincingly backed up with evidence.
1 September 2021
14 April 2021
23 February 2021
13 January 2021
Nothing matches.
Beliefs
Korrents What they believe 26 beliefs — each backed by an exact quote.
Each is a — compiled by korrents.com, not by them: the one-line wordings are korrents', the quotes are theirs.
Recent
Reliability in a data system means fault tolerance above all: the system keeps working when a network link is cut or a node dies.
for me, reliability means fault tolerance primarily. So, meaning that a system should, on the whole, continue working even if like a network link is interrupted or a node crashes or something like that.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
Scalability is not a property a system has; it is the set of mechanisms it has for coping with a change in load.
Scalability is just like what mechanisms we have for dealing with changes in load. If load increases, how can we add computing capacity to a system, for example, so that the system still continues working?
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
Scaling down is half of scalability: cost and capacity should track load in both directions, right down to a service that is nearly free to run.
generally, like you just want the the cost and the computing capacity to be roughly proportional to the load that you have. And at the low end, that means actually being able to scale down to something that is extremely cheap to run.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
Show 23 more
Building databases on object stores is a genuine architectural break, because replication moves out of the database and into the storage layer.
And now suddenly people are building databases on top of object stores, for example. And now the replication happens at the object store level, no longer at the database level.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
An engineer writing business logic loses nothing by not understanding the layer underneath, exactly as they lose nothing by not managing memory.
You're building higher-level business logic, actually I think it's just fine for people not to care about memory management.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
Higher-level abstractions do not abolish low-level knowledge; they concentrate it in the smaller group of people who build the abstractions.
But, somebody still has to build those lower-level abstractions and from lower-level components. Somebody's got to implement the cloud services.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
Knowing a little about how a storage engine works internally is a superpower for an application developer who will never build one.
And so those are those places where I feel like knowing a bit about the the internals is actually like a superpower.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
Europe being locked out of US cloud services is still unlikely, but it is no longer unthinkable, and serious systems should be engineered against it.
So, what if geopolitics was to go horribly wrong and tensions escalate and Europe find itself suddenly locked out of US cloud services? I hope that doesn't happen. I still think it's fairly unlikely, but it's no longer unthinkable.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
As AI writes more of the code, an engineer's job shifts away from expressing logic in a language and towards articulating high-level trade-offs.
Maybe as AI writes more and more code of our code, it's less about like the details of how you express logic in a particular programming language and much more about those kinds of high-level trade-offs.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
Sharding across machines has faded as a concern because hardware got more powerful, not because the cloud made it easy.
But, at least this sort of sharding across multiple machines is maybe become less of a pressing issue just because more and more workloads can just run on a single machine. Some people still have very large-scale workloads that do have to be sharded across multiple machines. So, it's not going away entirely.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
Replication earns its place even in a small system, because what it buys is fault tolerance rather than scale.
And uh replication is still relevant even at smaller scales because that's for fault tolerance, that's not for scalability.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
A distributed system must assume there is no upper bound whatsoever on how long a message can take to cross the network.
we just assume that there's no upper bound on how long it might take for a message to go over the network. So, you send a message, it might arrive within 100 microseconds, or it might take 10 years.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
Clocks are right most of the time, and that is precisely why a distributed system may never rely on them.
it's very easy to assume that your clocks are correct and most of the times the clocks are pretty correct but we just can't rely on it because actually they're just not precise enough uh on the whole.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
How much fault tolerance to buy is a business decision, not an engineering one; the engineer's job is to inform it and then step back.
And so, the goal of this chapter is really just to give people the information in order to make an educated decision. But I don't want to make that decision for people. That's for businesses themselves to decide.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
MapReduce is dead: nobody uses it any more, and it survives in the literature only as a teaching device for its successors.
In particular, for example, coverage of MapReduce was quite detailed in the first edition. But basically, MapReduce is dead. Nobody uses it anymore.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
If you got into technology to change the world, then thinking about what your technology does to the world is part of the job, not an optional extra.
people say that uh you get into tech in order to change the world. If you want to change the world, then thinking about the impact that your technologies have on the world is part of your job.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
A test checks a handful of examples; a proof settles an infinite state space, which is why formal methods can say things testing never can.
But a proof can reason about potentially infinite state spaces. So, it can tell you things about like every possible thing that could possibly happen in the entire universe.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
AI-written code makes formal proof necessary, because human review of all that generated code becomes the bottleneck.
But also LLMs increase the need for these formal proofs because, you know, we're live coding a bunch of stuff. If we have to manually review all of that code, then that will become the bottleneck.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
Formal verification is about to become economical, because models are getting good enough at writing the proofs that humans no longer have to.
One is that the LLMs are getting increasingly good at writing these proofs. And if we don't have to write the proof by hand as humans, it just becomes feasible to do them in situations where previously it would have not been economical.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
The reason a software-as-a-service business can charge a subscription is that it holds a gun to the customer's head over their own data.
software as a service businesses, for example the whole reason why they can charge a subscription is because they are able to essentially hold a gun to the customer's head and say, "Pay us at your subscription, otherwise we will delete all your data."
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
Power over data belongs with end users rather than cloud operators, and software should be built to move it back to them.
local-first software, which is this idea that we want to take away a bit of the power from cloud operators and give it back to end users. So, end users should be more in control of their own data and less dependent on cloud services for providing the applications and the data that that the users need.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
Work that runs against companies' commercial incentives will only get done by people whose income does not depend on those incentives.
there I feel like in academia, I have the freedom to work on things that go against this commercial incentive of companies and say like, "Actually, no, I'm going to do what I think is right for the users."
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
Timestamps cannot order access-control decisions in a decentralised system, because the user whose access was revoked can simply forge them.
if a revoked user still wants to, say, vandalize a document, they can just backdate their edits, give it an earlier timestamp. So, relying on clocks is absolutely useless here because people can forge the timestamps from those clocks and thereby then potentially undermine the access control mechanism.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
The essay a student hands in is not the point of education; the thought process that produced it is the entire desired outcome.
We don't ask the students to write essays because we love reading their amazing essays. We ask them to write essays because they we want them to go through a thought process, which helps them learn something.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
People in industry routinely short-circuit their reasoning, adopting what they heard in a conference talk instead of thinking it through from first principles.
Often people in industry I feel like sort of have short circuit reasoning. Like don't maybe don't quite reason something through from first principles but just like oh I heard this from a conference talk I'm just going to go with that
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
Writing is how the thinking gets done, so a writer who hands the words to a machine has given away the part that mattered.
It's more that for myself, the process of writing is the way how I figure things out. And figuring things out is really my goal here. So, I'm I'm trying to figure it out in my own heads, and for that I just have to write it myself.
Designing Data-intensive Applications with Martin Kleppmann Said 22 Apr 2026
What is a korrent?
A korrent is a belief a person has stated in their own words: one sentence stating the claim, backed by a quote and a source, kept at korrents.com.
Under a name here, the quoted block is what they actually said. The korrent beneath it is the claim those words support, in korrents' wording — tap it to see the record, its source, and who else holds it.
Nobody here wrote their own korrents. They are compiled from public statements, and a person can change their mind, which is recorded too.
Recommends rcmnd
rcmnd What they recommend 1 thing — each backed by an exact quote.
Compiled by rcmnd.app, not by them: the quotes are theirs, and each links to the post it was copied from.
Recent
Loved The Future of Fusion Energy
The book provides an excellent, detailed answer to these questions, and more. It’s the best kind of popular science book: you don’t need a physics degree to read it, but it doesn’t fob you off with oversimplified hand-waving either; all of the core arguments are convincingly backed up with evidence.
martin.kleppmann.com 3 Jan 2022
affiliate link
Where they publish
Blog martin.kleppmann.com His blog on distributed systems, CRDTs and local-first software. Has a feed.
Recent
- Announcing the Centre for Cryptographic Trust Infrastructure (CCTI) 7 Oct 2026 Today I am happy to announce a very exciting new research project that I’ve been working towards for years! We are establishing the Centre for Cryptographic Trust Infrastructure (CCTI), funded initially by a £2.4m grant…
- Prediction: AI will make formal verification go mainstream 8 Dec 2025 Much has been said about the effects that AI will have on software development, but there is an angle I haven’t seen talked about: I believe that AI will bring formal verification, which for decades has been a bit of a…
- Pudding: user discovery for anonymity networks 5 Jul 2024 I’d like to introduce an exciting new research paper I worked on! It’s about a system called Pudding, and it was presented by Ceren at the IEEE Symposium on Security and Privacy, one of the top academic conferences on c…
Show 7 more
- 2023 year in review 4 Jan 2024 A lot has happened in the last year, so I thought it would be good to write up a review. My biggest change in 2023 was that my wife and I had a baby! This has brought a mixture of joys and frustrations, but overall it h…
- Verifying distributed systems with Isabelle/HOL 12 Oct 2022 This post also appears on Larry Paulson’s blog. We use distributed systems every day in the form of internet services. These systems are very useful, but also challenging to implement because networks are unpredictable.…
- Book Review: The Future of Fusion Energy 3 Jan 2022 I give a five-star ⭐️⭐️⭐️⭐️⭐️ rating to the following book: Jason Parisi and Justin Ball. The Future of Fusion Energy. World Scientific, 2019. ISBN 978-1-78634-749-7. Available from Amazon US, Amazon UK, and many other…
- Several podcast interviews 1 Sept 2021 I regularly get asked to give interviews on the topics that I work on, especially for podcasts. To make them easier to find for anybody who’s interested, I thought I would make a list. They touch on a range of different…
- It's time to say goodbye to the GPL 14 Apr 2021 The trigger for this post is the reinstating of Richard Stallman, a very problematic character, to the board of the Free Software Foundation (FSF). I am appalled by this move, and join others in the call for his removal…
- Building the future of computing, with your help 23 Feb 2021 For the last five or six years, since I bid goodbye to the startup scene and Silicon Valley, I have been increasingly working in public. I have written a book, given around 100 talks (many of which are available on YouT…
- Decentralised content moderation 13 Jan 2021 Who is doing interesting work on decentralised content moderation? With Donald Trump suspended from Twitter and Facebook, and Parler kicked off AWS, there is renewed discussion about what sort of speech is acceptable on…
Link verified 20 Sept 2026. Recent items update automatically from the channel.
Bluesky @martin.kleppmann.com Their Bluesky account.
Recent
- I wrote a new first-year undergraduate course on databases (focussed on how to use query languages, not on DBMS internals). It covers SQL, relational algebra, JSON documents, Cypher, and Datalog. Notes are freely available here: 4 Nov 2026 cl.cam.ac.uk
- We're hiring a research engineer to work with myself, @inkandswitch.com, and an amazing interdisciplinary team on using cryptographic proofs and AI to verify physical-world (e.g. manufacturing) processes: 2 Nov 2026 inkandswitch.com
- My department is hiring a professor in AI/ML. It's a good place to work, please apply if you fit the role! 28 Oct 2026 cam.ac.uk
Show 17 more
- Very impressive CRDT optimisation work from @jzhao.xyz, improving upon Seph Gentle’s already very fast Eg-walker 28 Oct 2026 jzhao.xyz
- When DDIA 2nd ed first came out, there were problems with print quality, leading to a bunch of negative reviews on Amazon. I believe those issues have now been fixed for months, but the negative reviews persist. Anyone want to add a review to counterbalance them? 27 Oct 2026 amazon.com
- Using Claude to help write some lecture notes on databases, and it basically quoted my own book back to me (without me asking for it). I do know it's in the training data because I got letters about the Anthropic copyright settlement… still, feels pretty weird 18 Oct 2026
- The unreasonable effectiveness of stickers as a currency for rewarding 3-year-olds 14 Oct 2026
- My copy of this just arrived. Looking forward to digging into it! 13 Oct 2026
- The numbering of spaces in this Cambridge car park is truly a work of art: 1, 2, 4, 5, 6, 7, 3, 8, 9, 10, 11, 12, 13, 15, 14, 28, 27, 26, 25, 24, 23, 22, 21, 20, 19, 18, 17, 16, 36, 37, 38, 39, 40, 41, 31, 35, 34, 33, 32, 30, 29, 43, 42, 56, 55, 54, 53, 52, 51, 50, 49, 48, 47, 46, 45, 44 13 Oct 2026
- According to the IMF, AI hardware and related tech products now account for more than 10% of global trade in goods! 🤯 12 Oct 2026 thenextweb.com
- Our Centre for Cryptographic Trust Infrastructure (CCTI) is among the funded projects! 7 Oct 2026
- We just announced a big grant we got from @aria-research.bsky.social! We're aiming to use cryptographic protocols to allow mutually untrusting parties and their agents to prove to each other that they are telling the truth 7 Oct 2026 martin.kleppmann.com
- There once was an old lady who had spent her entire life prospecting for diamonds, with little success. Then, one day during her retirement, she made a discovery: a home microwave oven, with the right settings, could turn quartz into diamond! 7 Oct 2026
- Notion wrote a nice blog post on how they use CRDTs, including our Peritext algorithm for rich text! This must be one of the largest CRDT deployments out there? 28 Sept 2026 notion.com
- Now even featured on TV! 😍 27 Sept 2026 youtube.com
- A researcher I'm working with in Cambridge, @zazbrown.com, just used AI and Lean to prove the Affine Plank Conjecture, an open problem for the last 75 years! (To be precise, he proved a special case not proven before; the general conjecture is still open.) 25 Sept 2026 zazbrown.com
- I only just found this post by @joehellerstein.bsky.social from February on how waiting, ordering, and commitment in distributed systems are all ways of looking at the same underlying idea: coordination. Good post, matches with how I think about it 25 Sept 2026 jhellerstein.github.io
- There's a new Dinky Door in Cambridge! The cyclopunt – half cargo bike, half punt – best vehicle ever! 17 Sept 2026 dinkydoors.co.uk
- Interactive visualisation of the Raft consensus algorithm, linked with the pseudocode from my lectures — looks amazing! 2 Sept 2026
- Oh this looks amazing. Instant impulse purchase 1 Sept 2026
Link verified 23 Sept 2026. Recent items update automatically from the channel.
About the English under a post
Some people here publish in a language other than English. Where they do, this site shows a machine translation beneath the post, in this typeface — the site's own, not theirs.
The post itself is never changed, moved or hidden: what is set in the serif above is exactly what the person published, and it is what to quote them on. A translation can be wrong in ways that matter, especially about tone.
Only the post's own words are translated. A quoted post, a linked article and a belief on korrents.com are left in their original language.