Hi,
Let’s say a backup dump is taken onprem for a singlestore database with 180 partitions.
Can the backup dump be restored on singlestore cloud against a database with same number of partitions ?
Also, can it be restored on the cloud against a database with a different number of partitions ?
Yes to the first question, no to the second question.
If you ask SingleStore | Develop with SQrL, Your SingleStore Copilot. it will give you details and references.
Thanks Eric.
I have a followup question :
On Aug 7th, a backup dump was restored to the cloud in a database called Aug7_data.
On Aug 8th, another backup dump was restored to the cloud in a database called Aug8_data.
Is there a way to merge the data of the 2 cloud databases Aug7_data and Aug8_data into one single database Global_data without running any long running insert commands?
Is there a way to restore the Aug8_data as an append to Aug7_data ?
You’d have to restore both DBs separately. If you want to merge them, you can do INSERT…SELECT commands to combine the data into target tables. It could be one of the original databases (as a target for the 2nd DB table data), or a 3rd DB, depending on what you want.
If the tables are huge and you don’t want to try to run INSERT…SELECT to move an entire table from one place to another because you are worried it will take too long or run out of resources, then break it into logical batches. E.g. if the source has a date column and there are 30 days of data, write a loop in an SP and do 30 batches. You could even use a “progress helper” table to tell you where you are with each batch, so if the whole system dies and you have to restart the transfer process, you can pick up where you left off. You can use multi-statement transactions for this.