Question

write fonal project of 5 pages on Managing Database in ASM

Answer 1

Answer :

ASM is based on the principle that the database should manage storage instead of requiring an administrator to do it. ASM eliminates the need for you to directlymanage potentially thousands of Oracle database files. ASM groups the disks in your storage system into one or more disk groups.

Use of ASM:

• ASM is Oracle's recommended storage management solution that provides an alternative to conventional volume managers, file systems, and raw devices.
• ASM uses disk groups to store datafiles; an ASM disk group is a collection of disks that ASM manages as a unit.

Managing data base in ASM :

• In the chapters so far, we have discussed the various components of ASM and how they all work together to provide a solid storage management solution, taking advantage of Oracle metadata and providing performance benefits by storing data the way that Oracle databases need them.
• But how do the relational database management system (RDBMS) and the new storage manager work together? At this point, every reader must be wondering about the answer to this important question.
• In this chapter, let’s discuss the interaction between ASM and the RDBMS. We will also discuss the various interfaces to the ASM instance and how to migrate data into and from ASM storage.ASM supports both a single-instance version of the Oracle Database and a clustered version.
• Although from a strategic point of view ASM best fits a clustered configuration such as Real Application Clusters (RAC), it is not short of any features when it comes to implementing ASM in a single-instance database.
• So in this chapter, unless otherwise mentioned, all discussions apply to both types of configurations.

Interaction Between ASM and Database:

• An RDBMS instance is the standard Oracle instance and is the client of the ASM instance.

• The RDBMS-to-ASM communications are always intranode—that is, the RDBMS will not contact the remote ASM instance (in case of RAC) for servicing disk group requests. This is true until 12c ASM.

• It is important to note that RDBMS instances do not pass an I/O request to the ASM instance—that is, RDBMS instances do not proxy the I/O request through ASM.

• When an RDBMS instance opens an ASM file such as the system tablespace data file, ASM ships the file’s extent map to the RDBMS instance, which stores the extent’s map in its System Global Area (SGA). Given that the RDBMS instance has the extent map for the ASM file cached in the SGA, the RDBMS instance can process reads/writes directly from the disk without further intervention from the ASM instance.

• As is the case with raw devices, the I/O request size is dictated by the RDBMS instance, not by ASM. Therefore, if the RDBMS needs to read or flush an 8KB database block, it will issue an 8KB I/O, not a 1MB I/O.

• Users often ask whether there is any dependency between Automatic Segment Space Management (ASSM)–based data files and ASM. There is no dependency between ASM and ASSM (although their acronyms only differ by one extra letter).

• ASM manages how data files are laid out across disks (among other things) without looking at the data stored in those data files. ASSM deals with empty blocks that exist within a single segment (it replaces freelists and freelist groups).

• ASM does not care if the data files contain segments that are freelist-managed or autospace-segment-managed via Bitmap Managed Block (BMB), so no conversion of internal structures takes place.

• An active RDBMS instance that uses ASM storage can operate just like a typical database instance. All file access is directly performed with minimal ASM intervention.

• When a file is opened or created, the ASM instance sends the file layout to the RDBMS instance and all subsequent inputs/outputs (I/Os) are done using the extent map stored in the RDBMS instance.

• All ASM file access can only be done via the RDBMS instance and its utilities. For example, database backups of ASM files can be performed only with RMAN.

• Note that utilities such as the Unix dd command are not recommended for backing up or restoring ASM disk groups.

• Database instances usually interact with the ASM instance when files are created, resized, deleted, or opened.

• ASM and RDBMS instances also interact if the storage configuration changes, such as when disks are added, are dropped or fail.

• From a user perspective, the database-file-level access (read/write) of ASM files is similar to non-ASM, except that any database filename that begins with a plus sign (+) will automatically be handled and managed using the ASM code path. However, with ASM files, the database file access inherently has the characteristics of raw devices—that is, unbuffered (direct I/O) with kernelized asynchronous I/O (KAIO).

• Because the RDBMS instance issues I/Os against raw devices, there is no need to set the database parameter FILESYSTEMIO_OPTION unless network file system (NFS) volumes are used as ASM disks. Additionally, because ASYNCIO is enabled by default, no additional parameters are needed to leverage ASM.

• In fact, from database side, no database init.ora parameters are mandatory to support ASM.