The difference between
normal redundancy and high redundancy is in the number of failure groups required: A normalredundancy
disk group typically has two failure groups, and a high-redundancy disk group will
have at least three failure groups. A failure group in ASM would roughly correspond to a redo log
file group member using traditional Oracle datafile management. External redundancy requires
that the redundancy be provided by a mechanism other than ASM (for example, with a hardware
third-party RAID storage array). Alternatively, a disk group might contain a non-mirrored disk
volume that is used for a read-only tablespace that can easily be re-created if the disk volume fails.
ASM Instance
ASM requires a dedicated Oracle instance, typically on the same node as the database that is
using an ASM disk group. In an Oracle Real Application Clusters (RAC) environment, each node
in a RAC database has an ASM instance.
An ASM instance never mounts a database; it only coordinates the disk volumes for other
database instances. In addition, all database I/O from an instance goes directly to the disks in a
disk group. Disk group maintenance, however, is performed in the ASM instance; as a result, the
memory footprint needed to support an ASM instance can be as low as 64MB.
For more details on how to configure ASM for use with RAC, see Chapter 10.
Background Processes
Two new Oracle background processes exist in the ASM instance.
Pages:
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330