WEKO3
-
RootNode
アイテム
The Report of The Third Lisp Contest and The First Prolog Contest
https://ipsj.ixsq.nii.ac.jp/records/31327
https://ipsj.ixsq.nii.ac.jp/records/31327d5aabe85-69f5-4f7d-be23-d3bcead3010f
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
IPSJ-PRO85033004.pdf (2.5 MB)
|
Copyright (c) 1985 by the Information Processing Society of Japan
|
|
| オープンアクセス | ||
| Item type | SIG Technical Reports(1) | |||||||
|---|---|---|---|---|---|---|---|---|
| 公開日 | 1985-09-13 | |||||||
| タイトル | ||||||||
| タイトル | The Report of The Third Lisp Contest and The First Prolog Contest | |||||||
| タイトル | ||||||||
| 言語 | en | |||||||
| タイトル | The Report of The Third Lisp Contest and The First Prolog Contest | |||||||
| 言語 | ||||||||
| 言語 | jpn | |||||||
| 資源タイプ | ||||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_18gh | |||||||
| 資源タイプ | technical report | |||||||
| 著者所属 | ||||||||
| Musashino Electrical Communication Laboratories Nippon Telegraph and Telephone Corporation | ||||||||
| 著者所属(英) | ||||||||
| en | ||||||||
| Musashino Electrical Communication Laboratories Nippon Telegraph and Telephone Corporation | ||||||||
| 著者名 |
OkunoHiroshiG.
× OkunoHiroshiG.
|
|||||||
| 著者名(英) |
Hiroshi, G.Okuno
× Hiroshi, G.Okuno
|
|||||||
| 論文抄録 | ||||||||
| 内容記述タイプ | Other | |||||||
| 内容記述 | This is the report of the Third Lisp Contest and the First Prolog Contest. The benchmarks for these Contests were presented at the 28th meeting of WGSYM of IPSJ on June 15 1984 and the timing results were gathered till August 1985. The benchmarks are distributed more than 70 sites both in Japan and U.S.A. and Europe. However almost all the participants are in Japan. It is not meant that almost all the implementations are developed in Japan. The number of Lisp and Prolog implementations measured in these Contests are more than fifty: The Third Lisp Contest 37 implementations The First Prolog Contest 20 implementations This report covers various Lisp and Prolog implementations; from mainframes to micro computers; from general-purpose machines to dedicated Lisp and/or Prolog machines. The limit of the length of this report allows me to present only the total execution time and the summary of the profile of each system. I had to cut the profile of each system and the detailed information such as garbage collection time facilities and machine specifications. However I think that the data presented here will be sufficient to analyze each system because the ratio of garbage collection time to total execution time is very small except a few benchmarks. The timing results of Lisp benchmarks are shown in alphabetical order of the name of each Lisp system with the exception of the arrangement of the page. The order to provide the timing results is different from that of Lisp benchmarks; first the timing results for interpreted codes are provided and then those for compiled codes are shown. This is because the system whose data is shown for both interpreted codes and compiled codes is only DEC-10 PROLOG system. The comparison of Lisp and Prolog implementations is very difficult. The main reason why Lisp and Prolog are used as a pricipal programming language in Artificial Intelligence research is due to an excellent programming environment which each Lisp or Prolog implementation provides. And one of the key points that Lisp/Prolog can provide such a good programming environments is to use interpreter. This is why I proposed to gather timing data for interpreted codes. In this report we have no space to describe facilities and programming environments. Therefore the analysis of the implementation is left to the implementors and the readers. Of course the speed is very important factor for good programming environments. I'll publish the complete report of these Contests. If you measure the speed of a new system or the data listed here is updated please send the results to me by a floppy disk a magntic tape or electric mail. My network address is as follows: uucp: {titcca ccut kddlab riken tsuda}!nttmecl!ntt20!okuno {hplabs ihnp4 mcvax ukc}!kddlab!nttmecl!ntt20!okuno junet: okun@ntt20.ntt Any comment or suggestion on this report or benchmarking is welcome. It's not the end of The Third Lisp Contest and The First Prolog Contest but the start of the next Lisp and Prolog Contest. | |||||||
| 論文抄録(英) | ||||||||
| 内容記述タイプ | Other | |||||||
| 内容記述 | This is the report of the Third Lisp Contest and the First Prolog Contest. The benchmarks for these Contests were presented at the 28th meeting of WGSYM of IPSJ on June 15, 1984 and the timing results were gathered till August 1985. The benchmarks are distributed more than 70 sites both in Japan and U.S.A. and Europe. However, almost all the participants are in Japan. It is not meant that almost all the implementations are developed in Japan. The number of Lisp and Prolog implementations measured in these Contests are more than fifty: The Third Lisp Contest 37 implementations The First Prolog Contest 20 implementations This report covers various Lisp and Prolog implementations; from mainframes to micro computers; from general-purpose machines to dedicated Lisp and/or Prolog machines. The limit of the length of this report allows me to present only the total execution time and the summary of the profile of each system. I had to cut the profile of each system and the detailed information such as garbage collection time, facilities and machine specifications. However, I think that the data presented here will be sufficient to analyze each system, because the ratio of garbage collection time to total execution time is very small except a few benchmarks. The timing results of Lisp benchmarks are shown in alphabetical order of the name of each Lisp system with the exception of the arrangement of the page. The order to provide the timing results is different from that of Lisp benchmarks; first, the timing results for interpreted codes are provided, and then those for compiled codes are shown. This is because the system whose data is shown for both interpreted codes and compiled codes is only DEC-10 PROLOG system. The comparison of Lisp and Prolog implementations is very difficult. The main reason why Lisp and Prolog are used as a pricipal programming language in Artificial Intelligence research is due to an excellent programming environment which each Lisp or Prolog implementation provides. And one of the key points that Lisp/Prolog can provide such a good programming environments is to use interpreter. This is why I proposed to gather timing data for interpreted codes. In this report, we have no space to describe facilities and programming environments. Therefore, the analysis of the implementation is left to the implementors and the readers. Of course, the speed is very important factor for good programming environments. I'll publish the complete report of these Contests. If you measure the speed of a new system or the data listed here is updated, please send the results to me by a floppy disk, a magntic tape or electric mail. My network address is as follows: uucp: {titcca, ccut, kddlab, riken, tsuda}!nttmecl!ntt20!okuno {hplabs, ihnp4, mcvax, ukc}!kddlab!nttmecl!ntt20!okuno junet: okun@ntt20.ntt Any comment or suggestion on this report or benchmarking is welcome. It's not the end of The Third Lisp Contest and The First Prolog Contest, but the start of the next Lisp and Prolog Contest. | |||||||
| 書誌レコードID | ||||||||
| 収録物識別子タイプ | NCID | |||||||
| 収録物識別子 | AN10485570 | |||||||
| 書誌情報 |
情報処理学会研究報告プログラミング（PRO） 巻 1985, 号 30(1985-PRO-033), p. 1-24, 発行日 1985-09-13 |
|||||||
| Notice | ||||||||
| SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc. | ||||||||
| 出版者 | ||||||||
| 言語 | ja | |||||||
| 出版者 | 情報処理学会 | |||||||
