I've read a lot of articles about distributed Haskell. Much work has been done but seems to be in the area of distributing computations. I saw the remote package which seems to implement Erlang-style messaging passing but it is 0.1 and early stage.
I'd like to implement a system where there are many separate processes that provide distinct services, and are tied together by several main processes. This seems to be a natural fit for Erlang, but not so for Haskell. But I like Haskell's type safety.
Has there been any recent adoption of Erlang-style process management in Haskell?
If you want to learn more about the remote package, a.k.a CloudHaskell, see the paper as well as Jeff Epstein's thesis. It aims to provide precisely the actor abstraction you want, but as you say it is in the early stages. There is active discussion regarding improvements on the parallel-haskell mailing list, so if you have specific needs that remote doesn't provide, we'd be happy for you to jump in and help us decide its future directions.
More mature but lower-level than remote is the haskell-mpi package. If you stick to the Simple interface, messages can be sent containing arbitrary Serialize instances, but the abstraction is still way lower than remote.
There are some experimental systems, such as described in Implementing a High-level Distributed-Memory Parallel Haskell in Haskell (Patrick Maier and Phil Trinder, IFL 2011, can't find a pdf online). It blends a monad-par approach of deterministic dataflow parallelism with a limited ability to make the I-structures serializable over the network. These sorts of abstraction have promise for doing distributed computation, but since the focus is on computing purely-functional values rather than providing Erlang-style processes, they probably wouldn't be a good fit for your application.
Also, for completeness, I should point out the Haskell wiki page on cloud and HPC Haskell, which covers what I describe here, as well as the subsection on distributed Haskell, which seems in need of a refresh.
I frequently get the feeling that IPC and actors are an oversold feature. There are plenty of attractive messaging systems out there that have Haskell bindings e.g. MessagePack, 0MQ or Thrift. IMHO the only thing you have to add is proper addressing of processes and decide who/what is managing this addressing capability.
By the way: a number of coders adopt e.g. 0MQ into their Erlang environments, simply because it offers the possibility to structure messaging via message brokers rather then relying on pure process to process messaging in super scale.
In a "massively multicore world" I personally assume that shared memory approaches will eventually be outperforming messaging. Someone can then always come and argue with asynchrony of course. But already when you write that you want to "tie together" your processes by "several main processes" you in fact speak about synchronization. Also, you can of course challenge whether a single function, process or thread is the right level of parallelization.
In short: I would probably see whether MessagePack or 0MQ could fit my needs in Haskell and care for the rest in my code.
Related
I'm considering a new language to learn those days to be used in high performance computing on a cluster of computers we have, among those languages, I'm considering Haskell.
I have read some about Haskell, but still have questions about using Haskell in high performance and distributed computing, which the language is known for, but I read some debates about Haskell is not good for those type of systems due to laziness, I can summarize my questions in the following lines:
Haskell uses green threads, which is great for handling big number of concurrent connections, but what happens when one of tasks takes longer than average and blocks the rest, does the whole thread block (Node.js style), forward the next task to another processor/thread (Golang), use reductions technique (Erlang), which kicks the task out of processing context after a pre-determined number of ticks, or else?
In a distributed computing environment, what happens to lazily-evaluated functions, do they have to be forced strict?
If one function/module requires strict evaluation, but it depends on other lazy functions/modules, shall I modify the code of other functions/modules to make them strict as well, or the compiler will handle this to me and force everything in that chain to strict or lazy.
When processing a very large sequence of data, how does Haskell handle parallel processing, is it by following some kind of implicit map-reduce technique, or I have do it by myself.
Is there a clustering abstract in the language, that handles the computing power for me, that automatically forwards the next task to the free processor wherever it is, be it on the same computer or another computer in the same cluster.
How does Haskell ensure fair-share of work is evenly distributed to all the available cores on the same computer or on the available cluster.
GHC uses a pool of available work (called sparks) and a work-stealing system: when a thread runs out of work, it will look for work in the pool or on the work queues of other threads that it can steal.
There is no built-in support for distributed computing as there is in (say) Erlang. The semantics are whatever your implementation defines. There are existing implementations like Cloud Haskell that you can look at for examples.
Neither. Haskell will automatically do whatever work is necessary to provide a value that is demanded and no more.
Haskell (and GHC in particular) does not do anything to automatically parallelize evaluation because there is no known universal strategy for parallelizing that is strictly better than not parallelizing. See Why is there no implicit parallelism in Haskell? for more info.
No. See (2).
For the same machine, it uses the pool of sparks and the work-stealing system described above. There is no notion of "clustering".
For an overview of parallel and concurrent programming in Haskell, see the free book of the same name by Simon Marlow, a primary author of GHC's runtime system.
Multithreading
As far as SMP parallelism† is concerned, Haskell is very effective. It's not quite automatic, but the parallel library makes it really easy to parallelise just about anything. Because the sparks are so cheap, you can be pretty careless and just ask for lots of parallelism; the runtime will then know what to do.
Unlike in most other languages, it is not a big problem if you have highly branched data structures, tricky dynamic algorithms etc. – thanks to the purely functional paradigm, parallel Haskell never needs to worry about locks when accessed data is shared.
I think the biggest caveat is memory: GHC's garbage collector is not concurrent, and the functional style is rather allocation-happy.
Apart from that, it's possible to write programs that look like they're parallel, but really don't do any work at all but just start and immediately return because of laziness. Some testing and experience is still necessary. But laziness and parallelism are not incompatible; at least not if you make sure you have big enough “chunks” of strictness in it. And forcing something strict is largely trivial.
Simpler, common parallelism tasks (which could be expressed in a map-reduce manner, or the classic array-vector stuff – the ones which are also easy in many languages) can generally be handled even easier in Haskell with libraries that parallelise the data structures; the best-known of these is repa.
Distributed computing
There has been quite some work on Cloud Haskell, which is basically Erlang in library form. This kind of task is less straightforward: the idea of any explicit message sending is a bit against Haskell's grain, and many aspects of the workflow become more cumbersome if the language is so heavily focused on its strong static typing (which is in Haskell otherwise often a huge bonus that doesn't just improve safety and performance but also makes it easier to write).
I think it's not far off to use Haskell in a distributed concurrent manner, but we can't say it's mature in that role yet. For distributed concurrent tasks, Erlang itself is certainly the way to go.
Clusters
Honestly, Haskell won't help you much at all here. A cluster is of course in principle a special case of a distributed setup, so you could employ Cloud Haskell; but in practice the needs are very different. The HPC world today (and probably quite some time into the future) hinges on MPI, and though there is a bit of existing work on MPI bindings, I haven't found them usable, at least not just like that.
MPI is definitely also quite against Haskell's grain, what with it's FORTRAN-oriented array centrism, weird ways of handling types and so on. But unless you go nuts with Haskell's cool features (though often it is so tempting!) there is no reason you couldn't write typical number-crunching code also in Haskell. The only problem is again support/maturity, but it's a considerable problem; so for cluster computing I'd recommend C++, Python or Julia instead.
An interesting alternative is to generate MPI-parallelised C or C++ code from Haskell. Paraiso is one nice project that does this.
Pipe dreams
I have often though about what could be done to make the distributed computing feasible in idiomatic Haskell. In principle I believe laziness could be a big help there. The model I'd envision is to let all machines compute independently the same program, but make use of the fact that Haskell evaluation has generally no predetermined order. The order would be randomised on each machine. Also the runtime would track how long some computation branch took to complete, and how big the result is. If a result is deemed both expensive and compact enough to warrant it, it would then be broadcast to the other nodes, together with some suitable hash that would allow them to shortcut that computation.
Such a system would never be quite as efficient as a hand-optimised MPI application, but it could at least offer the same asymptotics in many cases. And it could handle vastly more complex algorithms with ease.
But again, that's totally just my vague hopes for the not-so-near future.
†You said concurrency (which isn't so much about computation as about interaction), but it seems your question is in essence about pure computations?
Should I learn scala and AKKA for a simulation project. Are these technologies a good fit / worth the investment? The task is to perform https://www.dropbox.com/s/3lby24y26wp60to/assignment.pdf?dl=0 an event-based simulation to simulate an IOT edge data center and implement some scheduling algorithms.
If yes, which libraries would you suggest? https://github.com/scalation/scalation does not seem to be a parallel library.
This is an opinion-based question. Should not be asked here.
Anyways, I'm going to try to give you some pointers. Akka is a generic framework: you can build anything from it, but nothing in particular is an immediate fit (ok, some things fit better than others, but still).
In your case, while Akka is a valid fit (actors = agents), I'd look more into specialized softwares for ABM (Agent based modeling), you can find a massive list here .
In particular, I recommend Netlogo: it's a bit counterintuitive in terms of syntax if you have never used something akin to Lisp or other immutable variables languages ("let" etc.), but once you get the hang of it it's very powerful for the effort required.
And, if you come from a CS background, it should be super easy for you (it's normally used by non-CS people in various fields, and it's designed to be easy).
I'm trying to visualize some simple automatic physical systems (such things as pendulum, robot arms,etc.) in Haskell.
Often those systems can be described by equations like
df/dt = c*f(t) + u(t)
where u(t) represents some kind of 'intelligent control'. Those systems look to fit very nicely in the Functional Reactive Programming paradigm.
So I grabbed the book "The Haskell School of Expression" by Paul Hudak,
and found that the domain specific language "FAL" (for Functional Animation Language) presented there actually works quite pleasently for my simple toy systems (although some functions, notably integrate, seemed to be a bit too lazy for an efficient use, but easily fixable).
My question is, what's the more mature, up-to-date, well-maintained, performance-tuned alternative for more advanced, or even practical applications today?
This wiki page lists several options for Haskell, but I'm not clear about the following respects:
The status of "reactive", the project from Conal Eliott who is (as I understand it) one of the inventers of this programming paradigm, looks a bit stale. I love his code, but maybe I should try other more up-to-date alternatives? What's the primary difference between them, in terms of syntax/performance/runtime-stability?
To quote from a survey in 2011, Section 6, "... FRP implementations are still not efficient enough or predictable enough in performance to be used effectively in domains which require latency guarantees ...". Alghough the survey suggests some interesting possible optimizations, given the fact that FRP is there for more than 15 years, I get the impression that this performance problem might be something very or even inherently difficult to solve at least within a few years. Is this true?
The same author of the survey talks about "time leaks" in his blog. Is the problem unique to FRP, or something we are generally having when programming in a pure, non-strict language? Have you ever found it just too difficult to stabilize an FRP-based system over time, if not performant enough?
Is this still a research level project? Are the people like plant engineers, robotics engineers, financial engineers, etc. actually using them (in whaterver language that suits their needs)?
Although I personally prefer a Haskell implementation, I'm open to other suggestions. For example, it would be particularly fun to have an Erlang implementation --- it would then be very easy to have an intelligent, adaptive, self-learning server process!
Right now there are mainly two practical Haskell libraries out there for functional reactive programming. Both are maintained by single persons, but are receiving code contributions from other Haskell programmers as well:
Netwire focusses on efficiency, flexibility and predictability. It has its own event paradigm and can be used in areas where traditional FRP does not work, including network services and complex simulations. Style: applicative and/or arrowized. Initial author and maintainer: Ertugrul Söylemez (this is me).
reactive-banana builds on the traditional FRP paradigm. While it is practical to use it also serves as ground for classic FRP research. Its main focus is on user interfaces and there is a ready-made interface to wx. Style: applicative. Initial author and maintainer: Heinrich Apfelmus.
You should try both of them, but depending on your application you will likely find one or the other to be a better fit.
For games, networking, robot control and simulations you will find Netwire to be useful. It comes with ready-made wires for those applications, including various useful differentials, integrals and lots of functionality for transparent event handling. For a tutorial visit the documentation of the Control.Wire module on the page I linked.
For graphical user interfaces currently your best choice is reactive-banana. It already has a wx interface (as a separate library reactive-banana-wx) and Heinrich blogs a lot about FRP in this context including code samples.
To answer your other questions: FRP isn't suitable in scenarios where you need real-time predictability. This is largely due to Haskell, but unfortunately FRP is difficult to realize in lower level languages. As soon as Haskell itself becomes real-time-ready, FRP will get there, too. Conceptually Netwire is ready for real-time applications.
Time leaks aren't really a problem anymore, because they are largely related to the monadic framework. Practical FRP implementations simply don't offer a monadic interface. Yampa has started this and Netwire and reactive-banana both build on that.
I know of no commercial or otherwise large scale projects using FRP right now. The libraries are ready, but I think the people aren't – yet.
Although there are some good answers already, I'm going to attempt to answer your specific questions.
reactive is not usable for serious projects, due to time leak problems. (see #3). The current library with the most similar design is reactive-banana, which was developed with reactive as an inspiration, and in discussion with Conal Elliott.
Although Haskell itself is inappropriate for hard real-time applications, it is possible to use Haskell for soft realtime applications in some cases. I'm not familiar with current research, but I don't believe this is an insurmountable problem. I suspect that either systems like Yampa, or code generation systems like Atom, are possibly the best approach to solving this.
A "time leak" is a problem specific to switchable FRP. The leak occurs when a system is unable to free old objects because it may need them if a switch were to occur at some point in the future. In addition to a memory leak (which can be quite severe), another consequence is that, when the switch occurs, the system must pause while the chain of old objects is traversed to generate current state.
Non-switchable frp libraries such as Yampa and older versions of reactive-banana don't suffer from time leaks. Switchable frp libraries generally employ one of two schemes: either they have a special "creation monad" in which FRP values are created, or they use an "aging" type parameter to limit the contexts in which switches can occur. elerea (and possibly netwire?) use the former, whereas recent reactive-banana and grapefruit use the latter.
By "switchable frp", I mean one which implements Conal's function switcher :: Behavior a -> Event (Behavior a) -> Behavior a, or identical semantics. This means that the shape of the network can dynamically switch as it's run.
This doesn't really contradict #ertes's statement about monadic interfaces: it turns out that providing a Monad instance for an Event makes time leaks possible, and with either of the above approaches it's no longer possible to define the equivalent Monad instances.
Finally, although there's still a lot of work remaining to be done with FRP, I think some of the newer platforms (reactive-banana, elerea, netwire) are stable and mature enough that you can build reliable code from them. But you may need to spend a lot of time learning the ins and outs in order to understand how to get good performance.
I'm going to list a couple of items in the Mono and .Net space and one from the Haskell space that I found not too long ago. I'll start with Haskell.
Elm - link
Its description as per its site:
Elm aims to make front-end web development more pleasant. It
introduces a new approach to GUI programming that corrects the
systemic problems of HTML, CSS, and JavaScript. Elm allows you to
quickly and easily work with visual layout, use the canvas, manage
complicated user input, and escape from callback hell.
It has its own variant of FRP. From playing with its examples it seems pretty mature.
Reactive Extensions - link
Description from its front page:
The Reactive Extensions (Rx) is a library for composing asynchronous
and event-based programs using observable sequences and LINQ-style
query operators. Using Rx, developers represent asynchronous data
streams with Observables, query asynchronous data streams using LINQ
operators, and parameterize the concurrency in the asynchronous data
streams using Schedulers. Simply put, Rx = Observables + LINQ +
Schedulers.
Reactive Extensions comes from MSFT and implements many excellent operators that simplify handling events. It was open sourced just a couple of days ago. It's very mature and used in production; in my opinion it would have been a nicer API for the Windows 8 APIs than the TPL-library provides; because observables can be both hot and cold and retried/merged etc, while tasks always represent hot or done computations that are either running, faulted or completed.
I've written server-side code using Rx for asynchronocity, but I must admit that writing functionally in C# can be a bit annoying. F# has a couple of wrappers, but it's been hard to track the API development, because the group is relatively closed and isn't promoted by MSFT like other projects are.
Its open sourcing came with the open sourcing of its IL-to-JS compiler, so it could probably work well with JavaScript or Elm.
You could probably bind F#/C#/JS/Haskell together very nicely using a message broker, like RabbitMQ and SocksJS.
Bling UI Toolkit - link
Description from its front page:
Bling is a C#-based library for easily programming images, animations,
interactions, and visualizations on Microsoft's WPF/.NET. Bling is
oriented towards design technologists, i.e., designers who sometimes
program, to aid in the rapid prototyping of rich UI design ideas.
Students, artists, researchers, and hobbyists will also find Bling
useful as a tool for quickly expressing ideas or visualizations.
Bling's APIs and constructs are optimized for the fast programming of
throw away code as opposed to the careful programming of production
code.
Complimentary LtU-article.
I've tested this, but not worked with it for a client project. It looks awesome, has nice C# operator overloading that form the bindings between values. It uses dependency properties in WPF/SL/(WinRT) as event sources. Its 3D animations work well on reasonable hardware. I would use this if I end up on a project in need for visualizations; probably porting it to Windows 8.
ReactiveUI - link
Paul Betts, previously at MSFT, now at Github, wrote that framework. I've worked with it pretty extensively and like the model. It's more decoupled than Blink (by its nature from using Rx and its abstractions) - making it easier to unit test code using it. The github git client for Windows is written in this.
Comments
The reactive model is performant enough for most performance-demanding applications. If you are thinking of hard real-time, I'd wager that most GC-languages have problems. Rx, ReactiveUI create some amount of small object that need to be GCed, because that's how subscriptions are created/disposed and intermediate values are progressed in the reactive "monad" of callbacks. In general on .Net I prefer reactive programming over task-based programming because callbacks are static (known at compile time, no allocation) while tasks are dynamically allocated (not known, all calls need an instance, garbage created) - and lambdas compile into compiler-generated classes.
Obviously C# and F# are strictly evaluated, so time-leak isn't a problem here. Same for JS. It can be a problem with replayable or cached observables though.
There's a plethora of paradigms and methods for concurrent programming in use today. Software transactional memory, actors, shared state concurrency, tuple spaces and many, many more.
What I find lacking, however, is a library of interesting test problems for concurrency. One well known example is the "Dining Philosophers Problem", which is neither a complex enough nor motivating nor realistic one. Then there are many parallel algorithms (matrix multiplication, rendering, general nested data parallelism) that just require distribution of work, but no real concurrency with communication between threads of execution.
So, can anyone point me to some interesting sets of problems that require real concurrency in an interactive, perhaps even distributed environment, that are simple enough to use as examples for concurrency paradigms? Ideally, I want to find a set of problems to serve as a "lackmus-test" for concurrency paradigms (or to highlight their differences, as every paradigm has its strengths and weaknesses).
Any help is much appreciated :)
I've previously considered this exact issue, having previously proposed some concurrent programing paradigms myself :p
The conclusion I reached then is that such a test set doesn't seem to really exist in a language-independent manner. While it might be helpful for it to exist, there seem to be some fairly good reasons it doesn't (to the best of my knowledge).
Most of the focus within concurrent programming tends to be on data parallelism, such that the same operation is applied in parallel to different pieces of the same data set. The kinds of task-level parallelism (i.e. different tasks being performed in parallel, possibly sharing data) that I think you're talking about is actually not done very much. I think this is because it's kinda hard. But I think it's also kinda hard because most problems don't lend themselves particularly well to this kind of concurrency. Describing a distributed system in terms of concurrency primitives may be helpful, but these systems tend to be decoupled such that there is a protocol (written or implied) moderating their communication. People tend not to think of these kinds of systems as obviously "concurrent" programming situations, even though they are when viewed within the right frame (i.e. considering the "client" and "server" as agents operating in parallel with synchronisation at some points).
The only places I think you could find some sources of inspiration would be within individual implementations. Erlang, Occam (and Occam-pi), Alice, CML, Concurrent Haskell etc all are likely to have small test corpuses, but both the problems and their implementations are going to be biased towards being implementable within a specific language (because they obviously are implementable within that language!). Perhaps you could also look to the communities working on multi-party session types, and various process calculi such as pi-calculus, CCS and CSP to see what kinds of systems they are using as example models. The idea of a standard language-agnostic set of problems for describing concurrent communicating systems is appealing, but somewhat elusive at this point, I think.
I'm about to tackle what I see as a hard problem, I think. I need to multi-thread a pipeline of producers and consumers.
So I want to start small. What are some practice problems, in varying levels of difficulty, that would be good for multi-threading practice? (And not contrived, impractical examples you see in books not dedicated to concurrency).
What books or references would you recommend that focus on concurrency and give in-depth problems and cases?
(I'd rather not focus on the problem I want to solve. I just want to ask for good references and sample problems. This would be more useful to other users. I'm not stuck on the problem.)
The little book of semaphores is a good free book. The author takes a unique approach of first asking a problem and then presenting hints before answering. The problems increase in difficulty level gradually, and the book isn't written for any language in particular but covers general multithreading concepts.
If you have enough time to invest I would recommend the book "Concurrency: State Models & Java Programs, 2nd Edition" by Jeff Magee and Jeff Kramer, John Wiley&Sons 2006
You can ignore the Java part if you are using some other language
There's a language used to model processes and concurrent processes called FSP. It needs some time and energy to be invested in order to be proficient in the language. There's a tool (LTSA, both are free and supported by an Eclipse plugin or stand alone app) which verifies your models and make you pretty shure that your model is correct from the standpoint of concurrent execution.
Translating this models to your language constructs is then just a question of programming technique and few design patterns.
Most text book problems, like readers-writers, producers-consumers or dinning philosophers are all illustrations of the mutex. I would prefer to model a prototype which is a simplistic approximation the bigger problem and go ahead.
I have some times seen situations where dead-lock avoidance is what is needed and dead-lock prevention measures are being used. It is always a good idea to analyse if Banker's algorithm would suit the case or not.
Completely ignoring your request, I'll suggest that you should look at SEDA (staged event driven architecture) as a way to think about setting up a multi-threaded pipeline of producers and consumers.
I'm not sure what you are looking for. But in real world enterprise situation, we usually use some kind of messaging framework when doing producers consumers stuff. Tipically in Java, that's JMS. And you can use the excellent Spring Framework to help you along.
If you're working with Java at all (and possibly even if you're not), you should definitely read Java Concurrency In Practice.
To be honest, many real-world multithreading programs are not doing much more than reading/writing some value (whether string or int) -- circular buffers (as a network connection might need), readers/writers of log files, etc.
In fact, I'd say that if you implement (or find) a solid (and generic) circular buffer, and then run all thread-to-thread communication through those buffers as the only contact point, that'll cover a very large portion of any multithread syncing you might need to do. (Unless you're working in a buzzword-compliant environment, and need to tack "enterprise", "messaging", or whatever onto the buzzword list... or you're writing a database or operating system.)
(Note that "circular buffer" is a fairly C-centric term, being rooted in the relatively direct manipulation of a block of memory. Python's Queue class implements the same basic principle in a list-centric way, and I'm sure that numerous other languages have conceptually similar constructs under slightly different names...)