Re: How batch processing works
От | Adrian Klaver |
---|---|
Тема | Re: How batch processing works |
Дата | |
Msg-id | 1dec4905-f6b7-4366-bdc0-fe1ce22bdacd@aklaver.com обсуждение исходный текст |
Ответ на | Re: How batch processing works (Lok P <loknath.73@gmail.com>) |
Список | pgsql-general |
On 10/4/24 1:05 PM, Lok P wrote: > > > On Mon, Sep 23, 2024 at 12:53 AM Peter J. Holzer <hjp-pgsql@hjp.at > <mailto:hjp-pgsql@hjp.at>> wrote: > > > > Thank you so much. > > I was expecting method-3(batch insert) to be the fastest or > atleast as you said > > perform with similar speed as method-2 (row by row insert with > batch commit) > > Oops, sorry! I wrote that the wrong way around. Method 3 is the fastest. > I guess I meant to write "method2 takes about twice as long as method3" > or something like that. > > > As in case of batch insert below is the fastest one as it inserts > multiple rows in one statement. Similarly I understand, Delete can be > batched as below. However, can you suggest how an Update can be batched > in a simple/generic fashion in JDBC for an input data stream with > multiple input values. As because for an update if we write as below , > it will just do one row update at a time? > > Update <table_name> SET column1=?, column2=?, column3=? where > <PK_Column>=? ; UPDATE table_name SET column1 = vals.text_val, column2=int_val FROM (VALUES (1, 'dog', 23),(2, 'cat', 44)) AS vals (id, text_val, int_val) where table_name.id = vals.id; > > > INSERT INTO <table_name> VALUES (1, 'a'), (2, 'a'),(3,'a'); > Delete from <table_name> where column_name in (<value1>, > <value2>,<value3>...); > > -- Adrian Klaver adrian.klaver@aklaver.com
В списке pgsql-general по дате отправления: