Benchmarks show that GHC 8.8 has significantly better performance than GHC 8.6 in many cases.
Use the following GHC options:
-O2
-fdicts-strict
-fspec-constr-recursive=16
-fmax-worker-args=16
In many cases fusion-plugin can improve performance by better stream
fusion. However, in some cases performance may also regress. Please note
that the fusion-plugin package works only for GHC >= 8.6.
-
Install the fusion-plugin package or add it to the
build-dependssection of your program in the cabal file. -
Use the following GHC option in addition to the options listed in the previous section:
-fplugin=Fusion.Plugin
At the very least -O -fdicts-strict compilation options are
required. If these options are not used, the program may exhibit memory
hog. For example, the following program, if compiled without an
optimization option, is known to hog memory:
main = S.drain $ S.concatMap S.fromList $ S.repeat []
-
-fdicts-strictis needed to avoid a GHC issue leading to memory leak in some cases. -
-fspec-constr-recursiveis needed for better stream fusion by enabling theSpecConstroptimization in more cases. -
-fmax-worker-argsis needed for better stream fusion by enabling theSpecConstroptimization in some important cases. -
-fplugin=Fusion.Pluginenables predictable stream fusion optimization in certain cases by helping the compiler inline internal bindings and therefore enabling case-of-case optimization. In some cases, especially in some fileio benchmarks, it can make a difference of 5-10x better performance.
Concurrency without a threaded runtime may be a bit more efficient. Do not use threaded runtime unless you really need multi-core parallelism. To get multi-core parallelism use the following GHC options:
-threaded -with-rtsopts "-N"
Use GHC 8.8 for best performance.
GHC 8.2.2 may hog memory and hang when building certain application using streamly (particularly the benchmark programs in the streamly package). Therefore we recommend avoiding using the GHC version 8.2.x.
If performance is below expectations:
- Look for inlining functions in the fast path
- Strictness annotations on data, specially the data used as accumulator in folds and scans, can help in improving performance.
- Strictness annotations on function arguments can help the compiler unbox constructors in certain cases, improving performance.
- Sometimes using
-XStrictextension can help improve performance, if so you may be missing some strictness annotations. - Use tail recursion for streaming data or for large loops