Re: Jsonb extraction very slow - Mailing list pgsql-general

From Merlin Moncure
Subject Re: Jsonb extraction very slow
Date
Msg-id CAHyXU0zBrUowa2C_LzeG7n-A3SOccs6KqCUWrzuN8rfg+oU=2A@mail.gmail.com
Whole thread Raw
In response to Re: Jsonb extraction very slow  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-general
On Thu, Aug 11, 2016 at 8:45 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Jim Nasby <Jim.Nasby@BlueTreble.com> writes:
>> I never dug into why. As Tom posited, decompression might explain the
>> time to get a single key out. Getting 10 keys instead of just 1 wasn't
>> 10x more expensive, but it was significantly more expensive than just
>> getting a single key.
>
> What were you doing to "get ten keys out"?  If those were ten separate
> JSON operators, they'd likely have done ten separate decompressions.
> You'd have saved something by having the TOAST data already fetched into
> shared buffers, but it'd still hardly be free.

Huh -- FWICT there is no way to pull N values from a jsonb with # of
items M for any value of N other than 1 or M with a single operation.

merlin


pgsql-general by date:

Previous
From: Merlin Moncure
Date:
Subject: Re: Any reasons for 'DO' statement not returning result?
Next
From: "Ilya Kazakevich"
Date:
Subject: Re: schema advice for event stream with tagging and filtering