Re: New Copy Formats - avro/orc/parquet

Поиск
Список
Период
Сортировка
От Nicolas Paris
Тема Re: New Copy Formats - avro/orc/parquet
Дата
Msg-id 20180211200012.2agrfocyaf42td5v@gmail.com
обсуждение исходный текст
Ответ на Re: New Copy Formats - avro/orc/parquet  (Andres Freund <andres@anarazel.de>)
Ответы Re: New Copy Formats - avro/orc/parquet  (Andres Freund <andres@anarazel.de>)
Список pgsql-general
> > That is true, but the question is how significant the overhead is. If
> > it's 50% then reducing it would make perfect sense. If it's 1% then no
> > one if going to be bothered by it.
> 
> I think it's pretty clear that it's going to be way way much more than
> 1%. 

Good news but not sure to anderstand why.

> It's trivial to construct cases where input parsing / output
> formatting takes the majority of the time. 

Binary -> ORC
        ^
    |
   PROGRAM parsing/output formating on the fly

> And a lot of that you're going to be able to avoid with binary formats.

Still the above diagram shows both parsing/formating step, isn't it ?





В списке pgsql-general по дате отправления:

Предыдущее
От: Andres Freund
Дата:
Сообщение: Re: New Copy Formats - avro/orc/parquet
Следующее
От: Andres Freund
Дата:
Сообщение: Re: New Copy Formats - avro/orc/parquet