Joe Armstrong wrote a paper for last year’s HOPL-III conference on the history of Erlang. For some reason, I didn’t get a paper copy of those proceedings, and was too busy to notice their absence. Fortunately Lambda the Ultimate picked it up and supplied links to the paper and the accompanying presentation. Digging into the history of something like Erlang is always fascinating, and Armstrong has done a good job of explaining how Erlang came to be.
Here are a bunch of quotes on topics that I found interesting. I’ve grouped them into categories, but searching the PDF of the paper shouldn’t be hard if you want to know where they originated.
Sources of inspiration
Those familiar with Prolog will not find it at all surprising that Erlang has its roots in Prolog (mostly due to implementation reasons). What I did find interesting was the origin/history/viewpoint on the concurrency model
The explanations of what Erlang was have changed with time:
1. 1986 â€“ Erlang is a declarative language with added concurrency.
2. 1995 â€“ Erlang is a functional language with added concurrency.
3. 2005 â€“ Erlang is a concurrent language consisting of communicating components where the components are written in a functional language.
Today we emphasize the concurrency.
Note that the word actor never appears in those descriptions. Indeed, the word actor does not appear in the paper at all. So for all the discussion about Erlang’s usage of the actor model, it appears that the Erlang folks independently duplicated many of the ideas for Hewitt’s Actors. I think that is kind of interesting.
Lisp and Smalltalk are cited as inspirations, but more for the implementation of the runtime than for any features in the language. I came away from the paper with the impression that Armstrong and his colleagues are not paradigm ideologues. They are trying to get the job done.
There is a huge emphasis on reliability throughout the paper, supporting Steve Vinoski’s remarks about Erlang. I’l just include a series of quotes, which you can interpret as you see fit:
Erlang was designed for writing concurrent programs that “run forever”
At an early stage we rejected any ideas of sharing resources between processes because of the difficulty of error handling. In many circumstances, error recovery is impossible if part of the data needed to perform the error recovery is located on a remote machine and if that remote machine has crashed.
In order to make systems reliable, we have to accept the extra cost of copying data structures between processes and always make sure that processes have enough data to continue by themselves if other processes crash
The key observation here is to note that the error-handling mechanisms were designed for building fault-tolerant systems, and not merely for protecting from program exceptions. You cannot build a fault-tolerant system if you only have one computer. The minimal conï¬guration for a fault-tolerant system has two computers. These must be conï¬gured so that both observe each other. If one of the computers crashes, then the other computer must take over whatever the ï¬rst computer was doing.
This means that the model for error handling is based on the idea of two computers that observe each other. Error detection and recovery is performed on the remote computer and not on the local computer.
Links in Erlang are provided to control error propagation paths for errors between processes.
It was about this time that we realized very clearly that shared data structures in a distributed system have terrible properties in the presence of failures. If a data structure is shared by two physical nodes and if one node fails, then failure recovery is often im-possible. The reason why Erlang shares no data structures and uses pure copying message passing is to sidestep all the nasty problems of ï¬guring out what to replicate and how to cope with failures in a distributed system.
In our world, we were worried by software failures where replication does not help.
Here are some quotes related the design criteria for Erlang.
Changing code on the fly was an initial key requirement
the notion that three properties of a programming language were central to the efï¬cient operation of a concurrent language or operating system. These were:
1) the time to create a process
2) the time to perform a context switch between two different processes
3) the time to copy a message between two processes
The performance of any highly-concurrent system is dominated by these three times.
One of the earliest design decisions in Erlang was to use a form of buffering selective receive
Pipes were rejected in favor of messages
In the concurrent logic programming languages, concurrency is implicit and extremely ï¬ne-grained. By comparison Erlang has explicit concurrency (via processes) and the processes are coarse-grained.
The ï¬nal strategy we adopted after experimenting with many different strategies was to use per-process stop-and-copy GC. The idea was that if we have many thousands of small processes then the time taken to garbage collect any individual process will be small.
Current systems run with tens to hundreds of thousands of processes and it seems that when you have such large numbers of processes, the effects of GC in an individual process are insigniï¬cant.
The BEAM compiler compiled Erlang programs to BEAM instructions.
This next series of quotes will probably make the pure functional language people shake their heads, but i think that it’s important to understand Erlang in contrast with pure functional languages.
Erlang is not a strict side-effect-free functional language but a concurrent language where what happens inside a process is described by a simple functional language.
Behaviors in Erlang can be thought of as parameterizable higher-order parallel processes.
… the status of Erlang as a fully ï¬‚edged member of the functional family is dubious. Erlang programs are not referentially transparent and there is no system for static type analysis of Erlang programs. Nor is it relational language. Sequential Erlang has a pure functional subset, but nobody can force the programmer to use this subset; indeed, there are often good reasons for not using it.
An Erlang system can be thought of as a communicating network of black boxes.
In the Erlang case, the language inside the black box just happens to be a small and rather easy to use functional language, which is more or less a historical accident caused by the implementation techniques used.
History and Usage
One thing that I was looking for in the paper was more details on how long Erlang had been around (besides before Java), how big the largest programs/systems were, and so forth. Here is what I found.
This history spans a twenty-year period…
(The history starts in 1986)
The largest ever system built in Erlang was the AXD301. At the time of writing, this system has 2.6 millions lines of Erlang code.
The AXD301 is written using distributed Erlang. It runs on a cluster using pairs of processors and is scalable up to 16 pairs of processors.
In the analysis of the AXD reported in , the AXD used 20 supervision trees, 122 client-server models, 36 event loggers and10 ï¬nite-state machines. All of this was programmed by a team of 60 programmers.
As regards reliability, the AXD301 has an observed nine-nines reliability â€”and a four-fold increase in productivity was observed for the development process .
The AXD 301 is circa 1998.
Perhaps the most exciting modern development is Erlang for multicore CPUs. In August 2006 the OTP group released Erlang for an SMP.
This corroborates something that David Pollak told me at the RedMonk unconference during CommunityOne, namely that SMP support in Erlang had not been there very long. Of course, Erlang was running on systems with 16 physical (pairs, no less) of processings in a distributed environment. So while the runtime might not be that mature on SMP, the overall runtime for concurrency is probably a bit more mature than that. Nonetheless, worthwhile to know the precise facts.
All in all, I found the paper to be a very worthwhile read – (and a nice change from my usual intake of blog posts and tweets). One of my pet peeves about the computer business is the lack of awareness of the history of the field. At least I’ve removed a bit of my own ignorance as relates to Erlang.
I find the Erlang platform interesting. I mean its principles and ideas. I think it managed to go that way because it stayed fairly unknown for so long and wasn’t never meant to be sold as an “enterprisey” solution like J2EE or .NET were. Had it been part of a more “let’s build a platform that’ll allow us to sell service afterwards” type of effort I think Erlang wouldn’t be where it is now.
I wonder though if the Erlang syntax is not a barrier to a larger adoption. Again probably this is not a desired goal from the Erlang developers anyway and thus is not a real issue. Indeed, more likely Erlang products will stay as described “An Erlang system can be thought of as a communicating network of black boxes”. We’ll see those boxes appearing in many areas where reliability and scalability are critical and applications will hook up in some ways with a more accessible from a developer’s, thus a company’s cash spent on salaries, point of view. In that kind of scenario Erlang systems will do just fine and its those plugged-in applications that’ll fail them. Interestingly ejabberd for instance could well be one of those most important Erlang applications out there in a near future considering it provides the strength of Erlang while being dead easy to hook into using XMPP.
A side note, for pythonistas out there, I believe that Kamaelia is a very interesting project as it provides some aspects of what Erlang can provide as library written in Python. I encourage developers to have a look at it.
Regarding SMP support – to maximise the use of an SMP box with Erlang is easy – start as many different Erlang Virtual Machines as there are processors…
Erlang has *always* been SMP…
But Erlang didn’t have support for multi-core (many concurrent execution threads per processor) until people started building them – and it got support for multi-core pretty sharply then… (ie the virtual machine became multi (operating system) threaded…
The first working SMP version of Erlang was described in 1998 (http://www.erlang.se/publications/xjobb/0089-hedqvist.pdf). Back in those days, SMP hardware was big and expensive, and not something that we could use in embedded telecoms systems. For that reason, no money was spent trying to productify it.
So it is true that the existing SMP Erlang implementation has not been around long, even though the concepts have matured over the last 10 years since the first prototype. We have found it very stable, and it’s already in use in commercial products.
You are right in observing that the development of Erlang has always been a very pragmatic effort.
I think (but haven’t yet verified) it’s a popular misconception that Erlang implements the Hewitt actor model. I’ve scoured Google and haven’t found a single reference to the actor model from Erlang folk like Joe Armstrong. It’s not mentioned once in his new Erlang book either. My hypothesis is this myth is allowed to persist because it’s “close enough” to the truth, it’s harmless, and it is helpful to have a more generic label than “Erlang-style concurrency” when they clone Erlang’s messaging semantics, as many now languages have.
Note that Armstrong says Erlang implements CSP (Communicating Sequential Processes), which is very close to the Actor Model but not quite. There’s a page on the c2 wiki comparing them (http://c2.com/cgi/wiki?ActorsModel).
I’ve been planning on e-mailing Joe Armstrong about this for a paper I was working on for the Erlang workshop, but life and illness have gotten in the way in recent weeks so I haven’t gotten around to it and might not make the submission deadline. 🙁
Pingback: Internet Alchemy » links for 2008-06-03
Pingback: The Cambrian Period of Concurrency at Ted Leung on the Air
Do note that there are more ways to skin a Symmetric Multiprocessor than running many threads inside a single VM. Distributed Erlang worked fairly well for those purposes too. (Personally, I think the issues and trade-offs are subtle and that there may well a case for going back to the era of 1 VM/core to exploit huge numbers of cores.)
Regarding the intellectual roots (or inspiration/requirements) of Erlang, don’t forget the Plex programming language.
Pingback: Erlang: Links, News and Resources (1) « Angel “Java” Lopez on Blog
Pingback: links for 2008-06-03 « Internet Alchemy