This occurs when SQL Server is waiting for I/O operations to finish that don’t read table or index rows from disk. It’s normal for a thread to enter a wait state as soon as it requests an I/O operation, but a problem may exist if the wait time is significant and your top waits list also includes at least one of the following:
The SQL Server error log may also contain a message that I/O requests are taking longer than 15 seconds to complete. You need to find out which operation is causing this, and where the bottleneck originates. It commonly occurs during long-running I/O-bound operations, including BACKUP, ALTER DATABASE, CREATE DATABASE or database autogrowth.
Disk avg. read time
Disk avg. write time
Avg. disk queue length
Buffer cache hit ratio
Buffer free pages
Buffer page life expectancy
Machine: memory used
As a last resort, review whether disk resources are adequate for your server load:
If your disk drivers and BIOS are out of date, consider upgrading.
Think about moving log files to a different disk, so they’re separate from database data files. See Storage Top 10 Best Practises.