As always, lists like these don’t include Joe Armstrong's PhD thesis “Making reliable distributed systems in the presence of software errors” - http://erlang.org/download/armstrong_thesis_2003.pdf

Are there any other works that should also be included that you know of?

No disrespect meant to Mr. Armstrong, but it's possible it's never listed because it's basically a textbook. It's 295 pages. The rough average of all papers in OP is like 15 pages.

His thesis is only slightly longer than typical PhD theses, I think.

Does anyone use Elixir/Erlang anymore?

WhatsApp backend in Erlang + used in telecoms

The "WhatsApp used Erlang" story was a decade ago, doubt it is true now.

I doubt they changed out Erlang/Elixir, as it is mature backbone infrastructure. It has run many telecoms for over 30 years without issue.

The problem is finding talent that knows both distributed systems, and Erlang/Elixir. But Erlang/OTP can replace a lot of expensive kludges. =3

But Erlang/OTP can replace a lot of expensive kludges.

If you can convince your employer of that. I'm working on a system today (and not for the first time) that could have significantly reduced complexity by switching to Erlang (or Elixir) and OTP. Amusingly, it's a glorified, in-house telecom system that was developed because they chose not to buy an existing solution.

Yes, but it is boring telecom projects... so most people aren't interested.

Scala is kind of a more modern alternative.

If doing a twitter like platform, elixir with Phoenix channels could handle around >20k users per host. Very few other options work for that use case. =3

goroutines?

Lamport is the author of more than half of these, it's crazy how influential he is in this space

In addition to making LaTex which has almost nothing to do with it

It's the thing that baffles me the most with those people, the Knuths and Lamports and Carmarcks, and whatnot.

It's not like they have light schedules - they do research, they have classes, they attendconference, and they have jobs, etc...

And some benefit from tenure, which give them time to deeply focus on some topics.

But I can't imagine how I would write tex in ten years in Pascal on archaic machines if I had all the time to myself.

Doing that in the middle of understanding more computer science than my eyes would ever read ?

What do they do with their time ? What do they don't do ? Are they able to skip the eating and pooping and sleeping and cleaning their house and picking up parcels at the post office and grocery shopping parts ?

Or a proto Unix being written in three weeks. Concretely, what are the days of those three weeks made of ? Especially since it's famously when the spouse was NOT there to handle the housekeeping...

I see not amount of "life hacks" and "atomic habits" and whatnot helping with that. But it's probably just as natural to them as biking - and you can't explain biking, I guess ?

My brother was something of a prodigy. Whatever it is that people like Carmack have, he had it too (and Carmack was one of his personal heros).

I would say with my brother it wasn't just that he was analytically extremely sharp, it's that he could focus it on things in a way that bordered on unhealthy obcession.

For example during undergrad he took a semester off because he burned himself out taking an insane course load, while also working on research as an undergrad.

Anyhow he spent that semester and the following summer at my parents house, and he spent months in basically the same pattern. Work on his research topic on the family PC until he needs to read a paper or think about something, at which point he'd roll the office chair over to the nearby couch and flop onto that. Every once in a while he'd disappear to his old bedroom to sleep for a few hours or take a shower or such.

And I mean he lived basically half a year like that, just relentlessly drilling into his research topic with every waking moment.

I'll add my stories to the 2 brother comments, in my prepa my best friend was by far the best in class. He did the 2 year program in chemistry in the first 2 months by himself. He arrived once in exams after a full night of re-reading the 3 LOTR books and still nailed the best grade effortlessly. And to his own admission, he was far from the most productive / smarter guys he knew.

I remember listening to him and a teacher, the teacher would give some anecdotes on what he's seen at the ENS. People reciting world chess championship games at night to fall asleep. Reading massive math books (1000+ pages) in an afternoon. Another story I heard in a different context from another teacher, math students were assigned a python project, after an introductory course. They never did any programming before. One student wrote 300 LoC in a single function and showed it to the teacher. It didn't have a main function, so the teacher asked him how he tested it. The student didn't know what he was talking about. They added what's needed to run the function and it worked.

So, it still begs the question... "How ?" Or, more precisely: "How the f... ?"

Clearly this goes beyond "Grit" [1], here, right ? (Those people do not seem like they have to coerce themselves into focusing that much.)

Is there a known (somewhat rigorous or not) study of those "super geniuses" ? I know that the "memory champions" are covered in "Moonwalking with Einstein" [2] ; but that seems like a different beast...

[1] https://www.simonandschuster.com/books/Grit/Angela-Duckworth... [2] https://www.goodreads.com/book/show/6346975-moonwalking-with...

This is not a bad list for sure. Here are some deeper cuts for those looking for something a bit less mainstream:

"The Maintenance of Duplicate Databases" https://datatracker.ietf.org/doc/html/rfc677 (AFAIK the genesis of the use of logical clocks in distributed systems).

"Chain Replication for Supporting High Throughput and Availability" https://www.usenix.org/legacy/event/osdi04/tech/full_papers/... (Chain replication is how a huge percentage of real-world cloud-scale data replication is done).

"Brewer’s Conjecture and the Feasibility of Consistent, Available, Partition-Tolerant Web Services" (The formalization of CAP, which caused a ton of very poor trade-off thinking in the decade that followed by defining Availability in a very goofy way. Still a classic.)

"Paxos Made Live" https://research.google/pubs/paxos-made-live-an-engineering-... (Brought a much-needed engineering perspective to a conversation that was largely theoretical up until this time.)

"Practical Byzantine fault tolerance" (Moved the conversation on Byzantine faults forward significantly).

This is just a short selection. There's so much good stuff going back in the 70s and 80s distributed database literature, for example (and in the modern systems and DB literature too).

Question from a recent CS grad here, who just learnt about CAP: what do you mean by “very poor trade-off thinking in the decade that followed”?

Chain replication is how a huge percentage of real-world cloud-scale data replication is done

Is this still true today? I know that Facebook does [1], but I don't have references for any other places which use it.

[1] https://engineering.fb.com/2022/05/04/data-infrastructure/de...

I am a Lamport admirer.

I gradually realized that Lamport is more of the godfather of distributed system than Hinton is to deep learning. Lamport is less prominent than Shannon is to information theory. Shannon is the closest to any title of "gold-like" figure to a scientific discipline of universal relevance in modern society.

Lamport specifically revealed a philosophical connections between computer systems and physics, in the parallel of distributed consensus to relativity theory. To me, the enlightenment is that, the relations between events happening in a distributed system, is more fundamental than their absolute ordering, thus the central role of an "observer". I haven't really analyzed if this realization is from Lamport's papers, or my general reading and thinking, but I am moderately confident that general readings are heavily influenced by Lamport's papers, or can be traced back to be compatible with Lamport's thinking. I have not seriously study if this connection is valid in depth, which might be another pure amateur speculation of mine.

One thing I think Lamport falls short is that his writing is not easy to read and understand. I unconciously feel that Lamport (and Dario from Anthropic) probably share a hidden sense of intellectual supriority grew from their own experience throughout their career. So their writing (and Dario's gospel) all share a unchangable sense of narration from their own delicate and graceful ideas, much less of faciliating the understanding to their audience. In this cateogry, Shannon is abosolutely superior in any measure, in his writing, ideas are so naturally presented, although the implications of the ideas remain elusive due to the inherent depth.

Also, among the 3 prominent figures of modern AI: Hinton/Bengio are more like Shannon, Lecun is closer to Lamport.

Enough random rambling. Lamport, as indicated by the outweight presence in this list, is no doubt the single most important scientist in distributed systems.

the relations between events happening in a distributed system, is more fundamental than their absolute ordering

The important thing in most distributed systems is having an order. Having a single observer serialize events as it receives them is so much more tractable than trying to use absolute order. Using absolute order requires very precise time synchronization which is hard; using absolute order requires knowing when you have received all the reports of events that already happened which is hard. Determining a designated observer isn't typically easy, but having it determine the order it observes events is easy. If two events happen at a similar time it's typically not a big deal which one is considered first as long as all nodes will agree on the result --- let the designated observer just pick the first one it sees works pretty well. If your report takes an unexpectedly long time to make it to the designated observer, then it won't be first and you'll deal.

Much better than trying to figure out unknowable questions of relativity. :P

absolute order requires very precise time synchronization which is hard

Presumably relativity is the reason precise time synchronization (and thus absolute ordering) is hard.

That forms a basis for the difficulty. But then there's additional layers of difficulty in real networks where the path between nodes is often asymmetric, and you may also observe that elapsed time (A -> B -> A) is sometimes greater than elapsed time (A -> C -> B -> C -> A) or (A -> C -> B -> A) or (A -> B -> C -> A)

Doesn't each node just need to know its exact path (or latency of same) to the time source each is synchronizing from? The path from node to node doesn't matter because we're ordering log entries from the timestamp of the receiving node.

Imagine you take three atomic clocks, synchronize them, and move each within exactly 1 meter of one of those nodes, directly connected in an identical manner. Relativistic effects are constant. If there are elevation differences, you factor it once and done. The problem with this setup is not relativity. It is quantum uncertainty (and various interference sources, thermal radiation, etc).

How do you know the latency of the path if you haven't synchronized the clocks...

You might be able to control the latency on a LAN, but once you have servers in different locations, good luck. (GPS helps a lot, of course...)

There is this command called 'ping' that tells you that. But in my example, I included the network topology as a known, static factor.

There is no such thing as a general order; if people accept this then the problem relaxes and we stop trying to design impossible things.

Nodes shouldn't be observers. End users should be the observers. The ordering of event only matters from the perspective of the "I'm adding this to my shopping cart and clicking the buy button and then it said I bought it" story being internally consistent. The observer here is the end user and the end user knows what happens-before relationships were.

Databases should flip this ordering inside out and let the observers (or proxies for them) report what they did in which order. And it just happens that protocols for doing this already exist: distributed tracing encodes these relations.

A meaningful global order doesn't exist and attempts to conjure one only conjure a meaningless one, or one that throws away a lot of turns trying to perfectly order everything from some arbitrary internal observer.

Hot take of the day:

Computer scientists are in denial about it, but CS is a branch of theoretical physics, not mathematics. You can point to this or that model of computation, such as lambda calculus or mu-recursive functions and try to claim its abstracted well beyond the particular laws of physics for some specific universe, but they all have some kind of rate limit built into them... and where does the motivation for this idea, that it takes something (time, space, work) to compute something ultimately come from? That's right - from underlying physics itself[1] - from the Bekenstein bound or Bremermann's limit or the like.

Even apparently non-physically-realizable models of computation like non-deterministic Turing machines are ultimately informed by and motivated by concepts in physics... otherwise they would just be examples of chmess[2] and of no interest to anyone. Computer science is of course somewhat abstracted from the details, but no more so than, say, thermodynamics, where concepts like entropy or Gibbs free energy can be studied in the abstract without reference to whether we are talking about a gas of non-interacting molecules or the spins of a bunch of electrons trapped in a lattice.

So, it's of no surprise whatsoever that the fundamental problems of distributed computing are ultimately the same as those found in the relativity of simultaneity[3]. You've all been studying the same things all along, just with different tools and at different levels of abstraction.

[1]: https://en.wikipedia.org/wiki/Limits_of_computation

[2]: https://link.springer.com/article/10.1007/s11245-006-0005-2

[3]: https://en.wikipedia.org/wiki/Relativity_of_simultaneity

Related hot take

Maybe thats why one of the not-so-bad ideas how to go about distributed systems came from a guy who was trained as a physicist and used to complain to his fellow programmers that “a lot of systems actually break the laws of physics”[1]?

In distributed systems there is no real shared state (imagine one machine in the USA another in Sweden) where is the shared state? In the middle of the Atlantic? - shared state breaks laws of physics. State changes are propagated at the speed of light - we always know how things were at a remote site not how they are now. What we know is what they last told us. If you make a software abstraction that ignores this fact you’ll be in trouble.[2]

[1]: “The Mess We’re In”, 2014 https://www.youtube.com/watch?v=lKXe3HUG2l4 [2]: https://news.ycombinator.com/item?id=19708900

And yet we have CRDTs.

A thought provoking leading question I have is (1) a space heater throws power into a coil and generates heat and nothing more in the world (2) an LLM into a GPU throws power into the GPU and generates heat and organized information.

Is (2) generating less heat so that the laws of thermodynamics account for the increased amount of "useful stuff" generated (the organized information)? Switch (2) for a bitcoin miner if you prefer.

Is the act of increasing order in the data, somehow costing some energy? Is there a fundamental minimum cost to flipping bits? Something about the switch of energy to matter and back, seems involved here.

I think the nature of computation is holding some secrets about the meaning of the universe, if there is any.

Some other classics (applied dist systems):

- DDB: https://www.allthingsdistributed.com/files/amazon-dynamo-sos...

- MapReduce: https://www.cs.princeton.edu/courses/archive/fall13/cos518/p...

- Spark/RDDs: https://www.usenix.org/system/files/conference/nsdi12/nsdi12...

- BigTable: https://static.googleusercontent.com/media/research.google.c...

Worth mentioning that Dynamo (the classic 2007 paper) and DynamoDB (the modern AWS product) have fairly little in common architecturally.

For DynamoDB, check out https://www.usenix.org/conference/atc22/presentation/elhemal... https://www.usenix.org/conference/atc23/presentation/idziore... and my analysis of the differences here https://brooker.co.za/blog/2025/08/15/dynamo-dynamodb-dsql.h...

Obligatory shout out to Morris and Kaashoek's class at MIT - they update the reading list every year, and it's always great: https://pdos.csail.mit.edu/6.824/

Some of these probably aren't exactly classics but I do love:

- rendezvous hashing / consistent hashing

- Hybrid logical clocks

- "Don’t Settle for Eventual: Scalable Causal Consistency for Wide-Area Storage with COPS"

- "Scaling Replicated State Machines with Compartmentalization" (much more recent but really one of my favorite papers)

I want to work on coding for a service large enough to involve distributed systems someday. I keep studying the theory, but still...

Try doing the assignments from MIT 6.5840 Distributed Systems course by Morris and Kaashoek.

Satoshi Nakamoto. 2008. Bitcoin: A Peer-to-Peer Electronic Cash System.

The greatest anonymous dude alive.