[Edit 2012:] 4a) What is Paul’s recommendation for using trace flag 1118? Everyone should turn it on, on all instances of SQL Server, from SQL Server 2000 onwards. There’s no down-side to having it turned on.

http://www.sqlskills.com/blogs/paul/a-sql-server-dba-myth-a-day-1230-tempdb-should-always-have-one-data-file-per-processor-core/

But there’s now even better guidance, and what I also recommend. At PASS in 2011, my good friend Bob Ward, who’s the top guy in SQL Product Support, espoused a new formula: if you have less than 8 cores, use #files = #cores. If you have more than 8 cores, use 8 files and if you’re seeing in-memory contention, add 4 more files at a time.

http://technet.microsoft.com/library/Cc966545

Use TF-1118. Under this trace flag SQL Server allocates full extents to each tempdb object, thereby eliminating the contention on SGAM page. This is done at the expense of some waste of disk space in tempdb. This trace flag has been available since SQL Server 2000. With improvements in tempdb object caching in SQL Server 2005, there should be significantly less contention in allocation structures. If you see contention in SGAM pages, you may want to use this trace flag. Cached tempdb objects may not always be available. For example, cached tempdb objects are destroyed when the query plan with which they are associated is recompiled or removed from the procedure cache.

Only SQLCHAR is allowed for fmt files if the file storage type is Char.

1 Data files that are stored in character format use char as the file storage type. Therefore, for character data files, SQLCHAR is the only data type that appears in a format file.

2 You cannot bulk import data into text, ntext, and image columns that have DEFAULT values.

When you bulk export data from an instance of SQL Server to a data file:

• You can always specify char as the file storage type.
• If you enter a file storage type that represents an invalid implicit conversion, bcp fails; for example, though you can specify int for smallint data, if you specify smallint for int data, overflow errors result.
• When noncharacter data types such as float, money, datetime, or int are stored as their database types, the data is written to the data file in the SQL Server native format.

Less amount is better for SQL server if you have a x64 machine and enough physical memory.

http://support.microsoft.com/kb/889654/en-us

http://blogs.msdn.com/b/buckwoody/archive/2010/06/29/the-windows-page-file-and-sql-server.aspx

http://www.brentozar.com/archive/2009/11/christian-bolton-on-sql-server-memory/

http://blogs.technet.com/b/askperf/archive/2007/12/14/what-is-the-page-file-for-anyway.aspx

http://www.brentozar.com/archive/2008/03/sql-server-2005-setup-checklist-part-1-before-the-install/

Veeam can truncate log base on some option.

http://www.veeam.com/vmware-backup/help-center/v6_1/hyperv/index.html?vmware_replica_guest_processing.htm

It make backup chain broken.
When they perform truncate log, you can see 2 of the log like this.

SQL VSS is normal behavior to create SnapShot on disk.
But, Veeam Backup and replication with truncate log isn’t good option to manage database.

I recommend that do not enable truncate log on Veeam replication manager.

select top 10 * 
  from msdb.dbo.backupset
 where is_snapshot=1

--It's due to SQL VSS.
--Program_Name : "Microsoft SQL Server VSS Writer"
--ex)
--BACKUP DATABASE [DBNAME] TO VIRTUAL_DEVICE='{4C5C20F9-FE32-4FCA-BE43-FFFFFFFFFFFF}1' WITH SNAPSHOT,BUFFERCOUNT=1,BLOCKSIZE=1024

select top 10 * 
  from msdb.dbo.backupmediafamily
 where physical_device_name='NUL'

--It's due to Veeam.
--Program_Name : "Veeam Backup and Replication"
--ex)
--BACKUP LOG [DBNAME] TO DISK = 'NUL'

http://msdn.microsoft.com/en-us/library/ms189786%28v=sql.100%29.aspx

This is one of the function to generate sequence GUID to reduce fragmentation.

But, it’s not grantee uniqueness.
It’s wrapped by windows function UuidCreateSequential, http://msdn.microsoft.com/en-us/library/aa379322%28VS.85%29.aspx.

Creates a GUID that is greater than any GUID previously generated by this function on a specified computer since Windows was started.
After restarting Windows, the GUID can start again from a lower range, but is still globally unique.
When a GUID column is used as a row identifier, using NEWSEQUENTIALID can be faster than using the NEWID function.
This is because the NEWID function causes random activity and uses fewer cached data pages. Using NEWSEQUENTIALID also helps to completely fill the data and index pages.

2. uniqueidentifier sorting.

http://sqlblog.com/blogs/alberto_ferrari/archive/2007/08/31/how-are-guids-sorted-by-sql-server.aspx

3. uniqueidentifier converting

SELECT CONVERT(uniqueidentifier, 0x000102030405060708090A0B0C0D0E0F)
UNION ALL
SELECT CONVERT(uniqueidentifier, 0x00000000000000000123456789ABCDEF)
UNION ALL
SELECT CONVERT(uniqueidentifier, 0x000000000123456789ABCDEF00000000)
UNION ALL
SELECT CONVERT(UNIQUEIDENTIFIER, '00000000-0123-4567-89AB-CDEF00000000')

Tuning, Consulting and all DB related questions.

Please select this one, “Do not truncate logs”.

——————————————————————————————————

Use the Truncation logs section to define the scenario of transaction log handing:

•Select Truncate logs on successful backup only if you want Veeam Backup & Replication to trigger truncation of logs only after the job is finished successfully. In this case, Veeam agent will wait for the replication to complete, and then will trigger truncation of transaction logs. If truncation of transaction logs is not possible for some reason, the logs will remain untouched in the VM guest OS till the next start of the Veeam agent.

•Select Truncate logs immediately if you want Veeam Backup & Replication to trigger truncation of logs in any case, no matter whether the job finishes successfully or fails.

•Select Do not truncate logs if you do not want Veeam Backup & Replication to truncate logs at all. This option is recommended if you are using another tool to perform guest-level replication and this tool maintains consistency of the database state. In such scenario, truncation of logs with Veeam Backup & Replication will break the guest-level replication chain and cause it to fall out of sync.

———————————————————————————————————————–

When I checked backup list, it’s truncate log like this.

BACKUP LOG xxxDB TO DISK = ‘NUL’

So, physical_device_name is presented as ‘NUL’.

select bs.*
	 , bf.*
  from msdb.dbo.dbo.backupset bs
  join msdb.dbo.[backupmediafamily] bf
    on bs.media_set_id = bf.media_set_id
 where 1=1
   and database_name='MyDB' and type='L'
 order by backup_start_date desc