diff options
Diffstat (limited to 'README.md')
| -rw-r--r-- | README.md | 20 |
1 files changed, 11 insertions, 9 deletions
@@ -6,18 +6,20 @@ Inpla is a multi-threaded parallel interpreter of interaction nets. Once you write programs for sequential execution, it works also in multi-threaded parallel execution. Each thread is managed on each CPU-core with POSIX-thread library. -* The current version is 0.8.2, released on **28 April 2022**. (See [Changelog.md](Changelog.md) for details.) +* The current version is 0.8.2-1, released on **1 May 2022**. (See [Changelog.md](Changelog.md) for details.) The below graph shows speed-up ratio to threads numbers for programs in the following benchmark table. In this version Quick and Merge sorts work for smaller lists whose elements are 260000, while it was 400000. This is because OCaml raises stack overflow error for over 260000 elements. As the result, the ratio looks not so good. For the results for 400000 elements (that are not so bad), which had been shown here, see the description of v0.8.2 in [Changelog.md](Changelog.md) + + + - | | Haskell | OCaml | SML | Python | Inpla8 | Inpla8r | | -------------- | :------: | :------: | :----: | :----: | :------: | :------: | -| ack(3,11) | [2.30](comparison/Haskell/ack3-11.hs) | [0.46](comparison/OCaml/ack3_11.ml) | [**0.43**](comparison/SML/ack3-11.sml) | [-](comparison/Python/ack3-11.py) | [0.83](comparison/Inpla/src/ack-stream_3-11.in) | [0.71](comparison/Inpla/src/ack-stream_3-11-reuse.in) | -| fib 38 | [1.60](comparison/Haskell/fib38.hs) | [**0.15**](comparison/OCaml/fib38.ml) | [0.27](comparison/SML/fib38.sml) | [8.88](comparison/Python/fib38.py) | [0.39](comparison/Inpla/src/fib-38.in) | [0.40](comparison/Inpla/src/fib-38-reuse.in) | -| bsort 40000 | [34.48](comparison/Haskell/bsort-40000.hs) | [34.62](comparison/OCaml/bsort40000.ml) | [12.06](comparison/SML/bsort-40000.sml) | [79.53](comparison/Python/bsort-40000.py) | [2.98](comparison/Inpla/src/bsort-40000.in) | [**2.48**](comparison/Inpla/src/bsort-40000-reuse.in) | -| isort 40000 | [12.54](comparison/Haskell/isort-40000.hs) | [7.45](comparison/OCaml/isort40000.ml) | [2.94](comparison/SML/isort-40000.sml) | [36.89](comparison/Python/isort-40000.py) | [**1.15**](comparison/Inpla/src/isort-40000.in) | [1.21](comparison/Inpla/src/isort-40000-reuse.in) | -| qsort 800000 | [1.56](comparison/Haskell/qsort-800000.hs) | [-](comparison/OCaml/qsort800000.ml) | [1.14](comparison/SML/qsort-800000.sml) | [97.37](comparison/Python/qsort-800000.py) | [0.66](comparison/Inpla/src/qsort-800000.in) | [**0.37**](comparison/Inpla/src/qsort-800000-reuse.in) | -| msort 800000 | [1.55](comparison/Haskell/msort-800000.hs) | [-](comparison/OCaml/msort800000.ml) | [1.04](comparison/SML/msort-800000.sml) | [100.04](comparison/Python/msort-800000.py) | [0.48](comparison/Inpla/src/msort-800000.in) | [**0.35**](comparison/Inpla/src/msort-800000-reuse.in) | +| ack(3,11) | [2.30](comparison/Haskell/ack3-11.hs) | [0.46](comparison/OCaml/ack3_11.ml) | [**0.43**](comparison/SML/ack3-11.sml) | [-](comparison/Python/ack3-11.py) | [0.84](comparison/Inpla/src/ack-stream_3-11.in) | [0.71](comparison/Inpla/src/ack-stream_3-11-reuse.in) | +| fib 38 | [1.60](comparison/Haskell/fib38.hs) | [**0.15**](comparison/OCaml/fib38.ml) | [0.27](comparison/SML/fib38.sml) | [8.88](comparison/Python/fib38.py) | [0.41](comparison/Inpla/src/fib-38.in) | [0.43](comparison/Inpla/src/fib-38-reuse.in) | +| bsort 40000 | [34.48](comparison/Haskell/bsort-40000.hs) | [34.62](comparison/OCaml/bsort40000.ml) | [12.06](comparison/SML/bsort-40000.sml) | [79.53](comparison/Python/bsort-40000.py) | [3.02](comparison/Inpla/src/bsort-40000.in) | [**2.52**](comparison/Inpla/src/bsort-40000-reuse.in) | +| isort 40000 | [12.54](comparison/Haskell/isort-40000.hs) | [7.45](comparison/OCaml/isort40000.ml) | [2.94](comparison/SML/isort-40000.sml) | [36.89](comparison/Python/isort-40000.py) | [**1.15**](comparison/Inpla/src/isort-40000.in) | [1.24](comparison/Inpla/src/isort-40000-reuse.in) | +| qsort 260000 | [0.34](comparison/Haskell/qsort-800000.hs) | [0.25](comparison/OCaml/qsort800000.ml) | [0.27](comparison/SML/qsort-800000.sml) | [10.40](comparison/Python/qsort-800000.py) | [0.15](comparison/Inpla/src/qsort-800000.in) | [**0.12**](comparison/Inpla/src/qsort-800000-reuse.in) | +| msort 260000 | [0.39](comparison/Haskell/msort-800000.hs) | [0.29](comparison/OCaml/msort800000.ml) | [0.26](comparison/SML/msort-800000.sml) | [10.96](comparison/Python/msort-800000.py) | [**0.15**](comparison/Inpla/src/msort-800000.in) | [**0.15**](comparison/Inpla/src/msort-800000-reuse.in) | - **Comparison in execution time** with other implementations: **Haskell** (GHC version 8.10.7), **OCaml** (ocamlopt, the native-code compiler, version 4.08.1), **Standard ML of New Jersey** v110.74 (interpreter mode) and **Python** 3.8.5 in execution time. @@ -29,7 +31,7 @@ Inpla is a multi-threaded parallel interpreter of interaction nets. Once you wri - "fib 38" is computation to get the 38th Fibonacci number. - - "bsort *n*", "isort *n*", "qsort *n*" and "msort *n*" are computation of bubble sort, insertion sort, quick sort and merge sort for random *n*-element lists, respectively, followed by a validation check process. The blanks in OCaml are due to the following error: `Fatal error: exception Stack_overflow` + - "bsort *n*", "isort *n*", "qsort *n*" and "msort *n*" are computation of bubble sort, insertion sort, quick sort and merge sort for random *n*-element lists, respectively, followed by a validation check process. |