Tag: bigfile

XTTS: Bigfile Tablespaces

What if the database you want to migrate has bigfile tablespaces? Bigfile tablespaces are very often used in very large databases. Do they influence on the migration using cross-platform transportable tablespaces and incremental backups?

What Is It?

First, let’s briefly recap what a bigfile tablespace is:

A bigfile tablespace is a tablespace with a single, but potentially very large (up to 4G blocks) data file. Traditional smallfile tablespaces, in contrast, can contain multiple data files, but the files cannot be as large.

The benefits of bigfile tablespaces are the following:

  • A bigfile tablespace with 8K blocks can contain a 32 terabyte data file…

Important to note is that a bigfile tablespace contains one big data file. That file can be huge.

What Do You Normally Do?

When RMAN backs up a database or a tablespace it will parallelize by using several channels – each channel will process one or more data files. Imagine a 32 TB data file (from a bigfile tablespace). One RMAN channel will be allocated and needs to process that data file alone. That will take a while!

To solve that multisection backup was implemented. That allows multiple channels to work on the same data file. You can enable multisection backup by using the clause section size.

Backup

The initial, level 0 backup of the source database is executed by the Perl script during the migration. The Perl script generates the RMAN backup commands on-the-fly, and it does not specify a section size. In addition, it is not possible to specify section size as a default channel configuration (see RMAN show all command).

So there is no way to enable the use of multisection backup. Currently, multisection backup is simply not supported by the Perl script.

Now What

If multisection backup is essential to your migration, I suggest you create a Service Request and ask for an enhancement. The more customers request it, the more likely it is that the feature will be added.

If you have bigfile tablespaces you have only one option. Use multiple Perl scripts. It is a good workaround if you have many bigfile tablespaces. The workaround probably won’t add much value if you only have one.

I have another blog post describing the use of multiple Perl scripts.

Other Blog Posts in This Series