Re: post-freeze damage control

Sorry, I've just noticed that the letter is shown incorrectly. I rewrote it below.

As I understand it, you talked about speeding up SeqScan by faster re-searching through the use of a hash table. At the same time, we have to build it before that, when there was the initial lookup tuples, right?

I found this information in the ExecEvalHashedScalarArrayOp function, and I assume you meant this function in your message, right?

But I couldn't find information, when you told about cycling a long-expression on each incoming tuple. Could you ask me what function you were talking about or maybe functionality? I saw ExecSeqScan function, but I didn't see it.

Hi! Unfortunately, I was not able to fully understand your message. Could you explain it to me please?

On 09.04.2024 16:20, Andrei Lepikhov wrote:

 
Moreover, it helps even SeqScan: attempting to find a value in the hashed array is much faster than cycling a long-expression on each incoming tuple.

As I understand it, you talked about speeding up SeqScan by faster re-searching through the use of a hash table. At the same time, we have to build it before that, when there was the initial lookup tuples, right?

I found this information in the ExecEvalHashedScalarArrayOp function, and I assume you meant this function in your message.

But I couldn't find information, when you told about cycling a long-expression on each incoming tuple. Could you ask me what function you were talking about or maybe functionality? I saw ExecSeqScan function, but I didn't see it.

-- 
Regards,
Alena Rybakina
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company

-- 
Regards,
Alena Rybakina
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company