Re: [Cuirass] Missing database indexes?

[swedebugia <swedebugia@riseup.net>]

Hi

On 2018-11-10 18:33, Ludovic Courtès wrote:

snip

> --8<---------------cut here---------------start------------->8---
> sqlite> CREATE INDEX Builds_index_evaluation ON Builds(evaluation);

snip

> Anything else we should do?

I woke up today with an idea. :)

Now on berlin it seems that all jobset are in one enormous table, correct?

How about creating a database/sqlite file for every specification and a 
table for every jobset? This would also solve the problem with 
overrunning id. (there will never be more ids than jobs in a single 
jobset in a table)

This means cuirass only work on one database per specification (now 
there are 5 specs on berlin)

Every jobset having its own table/file will lower the total number of 
rows, thus lower the computing cost of a table scan.

-- 
Cheers
Swedebugia