| Safe Haskell | None |
|---|---|
| Language | Haskell98 |
Control.Monad.Par
Description
The monad-par package provides a family of Par monads, for
speeding up pure computations using parallel processors. (for a similar
programming model for use with IO, see Control.Monad.Par.IO.)
The result of a given Par computation is always the same - i.e. it
is deterministic, but the computation may be performed more quickly
if there are processors available to share the work.
For example, the following program fragment computes the values of
(f x) and (g x) in parallel, and returns a pair of their results:
runPar $ do
fx <- spawnP (f x) -- start evaluating (f x)
gx <- spawnP (g x) -- start evaluating (g x)
a <- get fx -- wait for fx
b <- get gx -- wait for gx
return (a,b) -- return resultsPar can be used for specifying pure parallel computations in
which the order of the computation is not known beforehand.
The programmer specifies how information flows from one
part of the computation to another, but not the order in which
computations will be evaluated at runtime. Information flow is
described using "variables" called IVars, which support put and
get operations. For example, suppose you have a problem that
can be expressed as a network with four nodes, where b and c
require the value of a, and d requires the value of b and c:
a
/ \
b c
\ /
dThen you could express this in the Par monad like this:
runPar $ do
[a,b,c,d] <- sequence [new,new,new,new]
fork $ do x <- get a; put b (x+1)
fork $ do x <- get a; put c (x+2)
fork $ do x <- get b; y <- get c; put d (x+y)
fork $ do put a (3 :: Int)
get dThe result of the above computation is always 9. The get operation
waits until its input is available; multiple puts to the same
IVar are not allowed, and result in a runtime error. Values
stored in IVars are usually fully evaluated (although there are
ways provided to pass lazy values if necessary).
In the above example, b and c will be evaluated in parallel.
In practice the work involved at each node is too small here to see
the benefits of parallelism though: typically each node should
involve much more work. The granularity is completely under your
control - too small and the overhead of the Par monad will
outweigh any parallelism benefits, whereas if the nodes are too
large then there might not be enough parallelism to use all the
available processors.
Unlike Control.Parallel, in Control.Monad.Par parallelism is
not combined with laziness, so sharing and granulairty are
completely under the control of the programmer. New units of
parallel work are only created by fork and a few other
combinators.
The default implementation is based on a work-stealing scheduler that divides the work as evenly as possible between the available processors at runtime. Other schedulers are available that are based on different policies and have different performance characteristics. To use one of these other schedulers, just import its module instead of Control.Monad.Par:
For more information on the programming model, please see these sources:
- The wiki/tutorial (http://www.haskell.org/haskellwiki/Par_Monad:_A_Parallelism_Tutorial)
- The original paper (http://www.cs.indiana.edu/~rrnewton/papers/haskell2011_monad-par.pdf)
- Tutorial slides (http://community.haskell.org/~simonmar/slides/CUFP.pdf)
- Other slides: (http://www.cs.ox.ac.uk/ralf.hinze/WG2.8/28/slides/simon.pdf, http://www.cs.indiana.edu/~rrnewton/talks/2011_HaskellSymposium_ParMonad.pdf)
- data Par a
- runPar :: Par a -> a
- runParIO :: Par a -> IO a
- fork :: Par () -> Par ()
- data IVar a
- new :: Par (IVar a)
- newFull :: NFData a => a -> Par (IVar a)
- newFull_ :: a -> Par (IVar a)
- get :: IVar a -> Par a
- put :: NFData a => IVar a -> a -> Par ()
- put_ :: IVar a -> a -> Par ()
- spawn :: NFData a => Par a -> Par (IVar a)
- spawn_ :: Par a -> Par (IVar a)
- spawnP :: NFData a => a -> Par (IVar a)
- module Control.Monad.Par.Combinator
- class NFData a
The Par Monad
A version that avoids an internal unsafePerformIO for calling
contexts that are already in the IO monad.
forks a computation to happen in parallel. The forked
computation may exchange values with other computations using
IVars.
Communication: IVars
data IVar a
creates a new IVar
creates a new IVar that contains a value
creates a new IVar that contains a value (head-strict only)
read the value in a IVar. The get can only return when the
value has been written by a prior or parallel put to the same
IVar.
read the value in an IVar. get can only return when the
value has been written by a prior or parallel put to the same
IVar.
put :: NFData a => IVar a -> a -> Par ()
put a value into a IVar. Multiple puts to the same IVar
are not allowed, and result in a runtime error.
put fully evaluates its argument, which therefore must be an
instance of NFData. The idea is that this forces the work to
happen when we expect it, rather than being passed to the consumer
of the IVar and performed later, which often results in less
parallelism than expected.
Sometimes partial strictness is more appropriate: see put_.
put a value into a IVar. Multiple puts to the same IVar
are not allowed, and result in a runtime error.
put fully evaluates its argument, which therefore must be an
instance of NFData. The idea is that this forces the work to
happen when we expect it, rather than being passed to the consumer
of the IVar and performed later, which often results in less
parallelism than expected.
Sometimes partial strictness is more appropriate: see put_.
like put, but only head-strict rather than fully-strict.
Operations
Like fork, but returns a IVar that can be used to query the
result of the forked computataion. Therefore spawn provides futures or promises.
spawn p = do r <- new fork (p >>= put r) return r
Like spawn, but the result is only head-strict, not fully-strict.
Spawn a pure (rather than monadic) computation. Fully-strict.
spawnP = spawn . return
module Control.Monad.Par.Combinator
This module also reexports the Combinator library for backwards compatibility with version 0.1.
class NFData a
A class of types that can be fully evaluated.
Since: 1.1.0.0
Instances
| NFData Bool | |
| NFData Char | |
| NFData Double | |
| NFData Float | |
| NFData Int | |
| NFData Int8 | |
| NFData Int16 | |
| NFData Int32 | |
| NFData Int64 | |
| NFData Integer | |
| NFData Word | |
| NFData Word8 | |
| NFData Word16 | |
| NFData Word32 | |
| NFData Word64 | |
| NFData () | |
| NFData Version | Since: 1.3.0.0 |
| NFData a => NFData [a] | |
| (Integral a, NFData a) => NFData (Ratio a) | |
| NFData (Fixed a) | Since: 1.3.0.0 |
| (RealFloat a, NFData a) => NFData (Complex a) | |
| NFData a => NFData (Maybe a) | |
| NFData a => NFData (Set a) | |
| NFData (Vector a) | |
| NFData (Vector a) | |
| NFData (IVar a) | |
| NFData (IVar a) | |
| NFData (a -> b) | This instance is for convenience and consistency with Since: 1.3.0.0 |
| (NFData a, NFData b) => NFData (Either a b) | |
| (NFData a, NFData b) => NFData (a, b) | |
| (Ix a, NFData a, NFData b) => NFData (Array a b) | |
| (NFData k, NFData a) => NFData (Map k a) | |
| NFData (MVector s a) | |
| NFData (MVector s a) | |
| (NFData a, NFData b, NFData c) => NFData (a, b, c) | |
| (NFData a, NFData b, NFData c, NFData d) => NFData (a, b, c, d) | |
| (NFData a1, NFData a2, NFData a3, NFData a4, NFData a5) => NFData (a1, a2, a3, a4, a5) | |
| (NFData a1, NFData a2, NFData a3, NFData a4, NFData a5, NFData a6) => NFData (a1, a2, a3, a4, a5, a6) | |
| (NFData a1, NFData a2, NFData a3, NFData a4, NFData a5, NFData a6, NFData a7) => NFData (a1, a2, a3, a4, a5, a6, a7) | |
| (NFData a1, NFData a2, NFData a3, NFData a4, NFData a5, NFData a6, NFData a7, NFData a8) => NFData (a1, a2, a3, a4, a5, a6, a7, a8) | |
| (NFData a1, NFData a2, NFData a3, NFData a4, NFData a5, NFData a6, NFData a7, NFData a8, NFData a9) => NFData (a1, a2, a3, a4, a5, a6, a7, a8, a9) |
(0.3) Reexport NFData for fully-strict operators.