Writing About Our Experiences With Oracle Databases
Sunday February 5th 2012

Latest Topics

11gR2:Listener Startup Issues

In this blog post I will be discussing listener startup issues faced in 11gR2 RAC. I will be constantly updating this [Read More]

Get Upgrading: Upgrade to Enterprise Manager Cloud Control 12c

Get Upgrading: Upgrade to Enterprise Manager Cloud Control 12c

This post describes the steps to upgrade an existing 11g Enterprise Manager Grid Control (11.1.0.1.0) to 12c [Read More]

Cluster SSH tool Utility

Cluster SSH tool Utility

Many times you come across scenario’s when you wish to open multiple ssh windows and execute same commands.e.g [Read More]

11g: Multiple failed login attempt can block New Application connections

In Oracle database 11g, a new feature/enhancement has potential of bringing down the application due to multiple [Read More]

Optimizer Choosing Nested-Loop Joins Instead of Hash-Joins

In one of my databases, one application query suddenly started to pick Nested-Loop joins instead of Hash-Joins and took [Read More]

Question:Sizing ASM LUN

We received following question from Mr Orlando through our Contact Page

How about a discussion on how to size our LUNS to use with ASM. For instance we have a Dell MD3000 with 44x300GB HDDs (protected by RAID-10) and we’re planning to create LUNs of 2TB on the Storage, and deliver them to Windows 2003. for ASM to use what is common around the world? Just pass the 2TB Luns to Windows or maybe create several 500GB and give them to ASM. This is a RAC10g install, Oracle 10g Standard Edition.

Please find below our response to the question. We would also like to hear from others on their experiences and any other suggestions/opinions they may gave.

Our Response

As per 10gR2 ASM Best Practices document (available on ASM OTN Homepage), we can add Disk having size up to 2^32 Mb as ASM Disk (Minimum being 4 Mb). Therefore technically 2 Tb Disk/LUN can be added to the ASM Diskgroup but we would not recommend so.

Only advantage of having 2 Tb LUN would be reduced ASM Disks resulting in easier management. But there are some pitfalls too. We believe most of the problems will be seen during Disk Rebalance.

Suppose we have 3 TB of data on Two ASM Disks of 2 Tb each. Now we need to add new disk of 2Tb.In this case 1 Tb of data movement will happen and it will impact DB performance . Rebalance operation can also happen in case we are using Normal/High Redundancy and Disk failure occurs (due to Hardware problem or issues like header or block corruption), then failure group would be in Hung State. This needs to be corrected by adding a new Disk .

This would again require large amount of data movement (1.5 Tb in this example) and will take longer time to finish.

Lun size of 300-500 Gb should be fine for VLDB configuration.  Having separate diskgroup for DATA and FRA (Flash Recover Area) is also recommended approach. One more thing which can be tried out in case of multi-terabyte database is to change the Allocation Unit size for ASM Diskgroup.Each Allocation Unit consumes some amount of memory in ASM SGA for storing the metadata information. Very Large Databases (VLDB) which can have sizes in TeraBytes will have too many AU’s allocated and also memory requirement for ASM instance will also increase. You can read more about this in our previous post Allocation Unit and Extents In ASM

Share
Related Tags: , ,

Leave a Comment

*

Recent Comments

Saurabh Sood had this to say

Hi Pawan, We cannot force the writes to a specific disk. At least I am not aware of any such thing. Oracle will Read the post

Amit had this to say

hi Pawan, ASM_PREFERRED_READ_FAILURE_GROUPS is mainly used for RAC instances and that too stretch clusters i.e when Read the post

PAWAN had this to say

HI, i am newbie.... please guide me! My manager asked me to configure ASM in 11g with single instance Read the post

PAWAN had this to say

HI, i am newbie.... please guide me! My manager asked me to configure ASM in 11g with single instance Read the post

Saurabh Sood had this to say

@Ganesh: The global stats were up-to-date for both the databases, though I did not check the col stats. But the Read the post

More from category

10gR2 Silent Install with 11gr2 CRS fails

I was trying to perform a 10.2 silent install with 11gR2 CRS. While doing pre-checks installer failed with following [Read More]

10.2 CRS startup issue

Today I faced a strange issue with CRS  post host reboot. CRS was not coming up and we could see following message in [Read More]

Performance Management Guide on AIX

While trying to find the amount of physical memory used by oracle process on AIX, I got reference of a document from [Read More]

Effect Of Multiple SHMMAX Settings

Last week I saw a warning message at database startup time saying: WARNING: EINVAL creating segment of size [Read More]

Physical Corruption: ORA-1578 part 3

Recently we encountered a block corruption issue in 10g database on Linux x86 64 bit and using ASM for storing database [Read More]

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 4 other subscribers