Database Reference
In-Depth Information
fills, a log switch occurs and Oracle will begin to write to the next online redo log. If the data-
base is in ARCHIVELOG mode (see the section “NOARCHIVELOG and ARCHIVELOG
Modes”), the ARCH process will be responsible for taking that filled archived redo log and
copying it to one or more backup locations.
If the database is in ARCHIVELOG mode, then the ARCH process will start automati-
cally. Oracle can also start more than one ARCH process if multiple redo logs need to be
archived. For ARCH to work properly, you will need to configure the appropriate archiving
locations (see “Configuring the Database for Backup and Recovery” later in this chapter for
more information). The ARCH process is so vital to backup and recovery of the database
that if it cannot copy the archived redo logs to the mandatory archived log destinations, the
database will eventually stall until the problem is corrected.
Database Architecture
You might wonder why we spend time talking about the architecture of the database. We
don't want you to just understand how to do something; we want you to understand why
you do it. Understanding backup and recovery requires knowing the architecture of the Ora-
cle Database. This knowledge may well make the difference when you face a question and
you are not sure of the answer. If you understand how something works, then figuring out
an answer to a question becomes easier because you also know how it does not work.
If you are unsure of an answer to an OCP exam question, try to work out which answers
are not correct and eliminate them. By knowing how the architecture works, you will be
able to eliminate questions more easily and find the correct answer more often.
Multithreaded Oracle
Oracle Database 12 c introduced an optional multithreaded process/threaded model for
operating systems such as Unix and Linux. In this model, most of the database back-
ground processes are spawned as threads of a few primary processes. The Oracle multi-
threaded model should not be confused with Oracle Shared Servers. Shared Servers is a
user connection pooling mechanism, whereas multithreaded Oracle is a different way of
spawning the processes that make Oracle run.
You can enable multithreaded Oracle through the use of the THREADED_EXECUTION param-
eter. When this parameter is set to TRUE , there will no longer be a one-to-one relationship
between an Oracle background process and an operating system process.
User Processes
User processes are the processes that are started by client programs to connect to the
database. In the case of local user processes (also called foreground processes to distin-
guish them from background database instance processes) on the database server, they
Search WWH ::




Custom Search