Skip to content

Commit

Permalink
documentation updates
Browse files Browse the repository at this point in the history
- issue #853
- pg_strom.enable_numeric_aggfuncs
  • Loading branch information
kaigai committed Nov 16, 2024
1 parent ddd181a commit 493f3f1
Show file tree
Hide file tree
Showing 13 changed files with 12 additions and 12 deletions.
2 changes: 1 addition & 1 deletion docs/arrow_fdw/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -394,7 +394,7 @@ <h2 id="arrow_fdw-virtual-column">Arrow_Fdw Virtual Column</h2>
<p>In Arrow_Fdw, you can refer to part of the file name as a column by using the foreign table option <code>pattern</code>. This is called a virtual column and is configured as follows.</p>
<pre><code>=# IMPORT FOREIGN SCHEMA f_lineorder
FROM SERVER arrow_fdw INTO public
OPTIONS (dir '/opt/arrow/mydata', pattern 'f_lineorder__${shipping}.arrow');
OPTIONS (dir '/opt/arrow/mydata', pattern 'f_lineorder_@{year}_${shipping}.arrow');
IMPORT FOREIGN SCHEMA

=# \d f_lineorder
Expand Down
2 changes: 1 addition & 1 deletion docs/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -250,5 +250,5 @@ <h2 id="support-policy">Support Policy</h2>

<!--
MkDocs version : 1.6.0
Build Date UTC : 2024-10-30 06:01:17.219408+00:00
Build Date UTC : 2024-11-16 06:20:09.721321+00:00
-->
2 changes: 1 addition & 1 deletion docs/ja/arrow_fdw/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -394,7 +394,7 @@ <h2 id="arrow_fdw">Arrow_Fdwの仮想列</h2>
<p>Arrow_Fdwでは、外部テーブルオプション<code>pattern</code>を使用する事でファイル名の一部を列として参照する事ができます。これを仮想列と呼び、以下のように設定します。</p>
<pre><code>=# IMPORT FOREIGN SCHEMA f_lineorder
FROM SERVER arrow_fdw INTO public
OPTIONS (dir '/opt/arrow/mydata', pattern 'f_lineorder__${shipping}.arrow');
OPTIONS (dir '/opt/arrow/mydata', pattern 'f_lineorder_@{year}_${shipping}.arrow');
IMPORT FOREIGN SCHEMA

=# \d f_lineorder
Expand Down
2 changes: 1 addition & 1 deletion docs/ja/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -242,5 +242,5 @@ <h2 id="_6">サポートポリシー</h2>

<!--
MkDocs version : 1.6.0
Build Date UTC : 2024-10-30 06:01:17.807183+00:00
Build Date UTC : 2024-11-16 06:20:10.295067+00:00
-->
2 changes: 1 addition & 1 deletion docs/ja/ref_params/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -161,7 +161,7 @@ <h2 id="_1">機能単位の有効化/無効化</h2>
<dd>GpuPreAggによる集約処理を有効化/無効化する。</dd>
<dt><code>pg_strom.enable_numeric_aggfuncs</code> [型: <code>bool</code> / 初期値: <code>on]</code></dt>
<dd><code>numeric</code>データ型を引数に取る集約演算をGPUで処理するかどうかを制御する。</dd>
<dd>GPUでの集約演算において<code>numeric</code>データ型は倍精度浮動小数点数にマッピングされるため、計算誤差にセンシティブな用途の場合は、この設定値を <code>off</code> にしてCPUで集約演算を実行し、計算誤差の発生を抑えることができます</dd>
<dd>GPUでの集約演算において<code>numeric</code>データ型は128bit固定小数点変数にマッピングされるため、極端に大きな数、あるいは高精度な数の集計を行う場合はエラーとなってしまいます。そのようなワークロードに対しては、この設定値を <code>off</code> にしてCPUで集約演算を実行するよう強制する事ができます</dd>
<dt><code>pg_strom.enable_brin</code> [型: <code>bool</code> / 初期値: <code>on]</code></dt>
<dd>BRINインデックスを使ったテーブルスキャンを有効化/無効化する。</dd>
<dt><code>pg_strom.cpu_fallback</code> [型: <code>enum</code> / 初期値: <code>notice</code>]</dt>
Expand Down
2 changes: 1 addition & 1 deletion docs/ja/search/search_index.json

Large diffs are not rendered by default.

Binary file modified docs/ja/sitemap.xml.gz
Binary file not shown.
2 changes: 1 addition & 1 deletion docs/ref_params/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -161,7 +161,7 @@ <h2 id="enablesdisables-a-particular-feature">Enables/disables a particular feat
<dd>Enables/disables GpuPreAgg</dd>
<dt><code>pg_strom.enable_numeric_aggfuncs</code> [type: <code>bool</code> / default: <code>on]</code></dt>
<dd>Enables/disables support of aggregate function that takes <code>numeric</code> data type.</dd>
<dd>Note that aggregated function at GPU mapps <code>numeric</code> data type to double precision floating point values. So, if you are sensitive to calculation errors, you can turn off this configuration to suppress the calculation errors by the operations on CPU.</dd>
<dd>Note that aggregated function at GPU mapps <code>numeric</code> data type to 128bit fixed-point variable, so it raises an error if you run the aggregate functions with extremely large or highly-precise values. You can turn off this configuration to enforce the aggregate functions being operated by CPU.</dd>
<dt><code>pg_strom.enable_brin</code> [type: <code>bool</code> / default: <code>on]</code></dt>
<dd>Enables/disables BRIN index support on tables scan</dd>
<dt><code>pg_strom.cpu_fallback</code> [type: <code>enum</code> / default: <code>notice</code>]</dt>
Expand Down
2 changes: 1 addition & 1 deletion docs/search/search_index.json

Large diffs are not rendered by default.

Binary file modified docs/sitemap.xml.gz
Binary file not shown.
2 changes: 1 addition & 1 deletion man/arrow_fdw.md
Original file line number Diff line number Diff line change
Expand Up @@ -472,7 +472,7 @@ In Arrow_Fdw, you can refer to part of the file name as a column by using the fo
```
=# IMPORT FOREIGN SCHEMA f_lineorder
FROM SERVER arrow_fdw INTO public
OPTIONS (dir '/opt/arrow/mydata', pattern 'f_lineorder_@{year}_${shipping}.arrow');
OPTIONS (dir '/opt/arrow/mydata', pattern 'f_lineorder_@\{year}_${shipping}.arrow');
IMPORT FOREIGN SCHEMA
=# \d f_lineorder
Expand Down
4 changes: 2 additions & 2 deletions man/ref_params.md
Original file line number Diff line number Diff line change
Expand Up @@ -69,12 +69,12 @@ This session introduces PG-Strom's configuration parameters.
@ja{
`pg_strom.enable_numeric_aggfuncs` [型: `bool` / 初期値: `on]`
: `numeric`データ型を引数に取る集約演算をGPUで処理するかどうかを制御する。
: GPUでの集約演算において`numeric`データ型は倍精度浮動小数点数にマッピングされるため、計算誤差にセンシティブな用途の場合は、この設定値を `off` にしてCPUで集約演算を実行し、計算誤差の発生を抑えることができます
: GPUでの集約演算において`numeric`データ型は128bit固定小数点変数にマッピングされるため、極端に大きな数、あるいは高精度な数の集計を行う場合はエラーとなってしまいます。そのようなワークロードに対しては、この設定値を `off` にしてCPUで集約演算を実行するよう強制する事ができます
}
@en{
`pg_strom.enable_numeric_aggfuncs` [type: `bool` / default: `on]`
: Enables/disables support of aggregate function that takes `numeric` data type.
: Note that aggregated function at GPU mapps `numeric` data type to double precision floating point values. So, if you are sensitive to calculation errors, you can turn off this configuration to suppress the calculation errors by the operations on CPU.
: Note that aggregated function at GPU mapps `numeric` data type to 128bit fixed-point variable, so it raises an error if you run the aggregate functions with extremely large or highly-precise values. You can turn off this configuration to enforce the aggregate functions being operated by CPU.
}

@ja{
Expand Down
2 changes: 1 addition & 1 deletion man/release_v5.3.md
Original file line number Diff line number Diff line change
Expand Up @@ -11,7 +11,7 @@ PG-Strom v5.3における主要な変更は点は以下の通りです。

- GpuJoin Partitioned Pinned Inner Buffer
- Arrow_Fdw Virtual Columns

- AVG(numeric), SUM(numeric)の誤差問題


- GpuJoin Pinned Inner Buffer
Expand Down

0 comments on commit 493f3f1

Please sign in to comment.