SQL Server IO size

sakuraime 2,321 Reputation points
2020-08-24T03:11:30.007+00:00

When sql server choose the I/O size for a single read/write (Data file random IO)?

8KB
64KB

or even larger

SQL Server
SQL Server
A family of Microsoft relational database management and analysis systems for e-commerce, line-of-business, and data warehousing solutions.
13,188 questions
0 comments No comments
{count} votes

9 answers

Sort by: Most helpful
  1. sakuraime 2,321 Reputation points
    2020-08-28T16:52:39.153+00:00

    " Different SQL Server operations will be different of reads and writes."

    any official MSDN doc for this ?

    0 comments No comments

  2. sakuraime 2,321 Reputation points
    2020-08-30T09:49:02.66+00:00

    I still can't see any official taking about that


  3. Shashank Singh 6,246 Reputation points
    2020-08-31T08:45:54.857+00:00

    This link has official downloadable white paper. I quote from the white paper

    Readahead pages/sec Pages reads/sec measures the number of physical page reads being issued per second.
    Read-ahead pages/sec measures the number of physical page reads that are performed using the SQL Server read-ahead mechanism. Read-ahead operations are used by SQL Server for scan activity (which is common for data warehouse and decision support workloads). These can vary in size in any multiple of 8 KB, from 8 KB through 512 KB. This counter is a subset of Pages reads/sec and can be useful in determining how much I/O is generated by scans as opposed to seeks in mixed workload environments.

    0 comments No comments

  4. Norm C 6 Reputation points
    2020-09-02T20:14:52.847+00:00

    When sql server choose the I/O size for a single read/write (Data file random IO)?

    8KB
    64KB


    You should work with your system or storage administrator, and depending on the type of system.. do they we still the term OLAP and OLTP ? If it's a highly transactional system you can actually design your storage to optimize for certain block sizes (smaller), where as for reporting and ETL systems you could optimize for throughput reads..so you can larger chunks (larger block sizes) defined for storage. You can do some research to help verify that, but that's my experience/research before and recall from optimizing storage related to dbs. Random IO definitely smaller sizes for defining your storage sub system.

    0 comments No comments