Why should you care?

When a report is slow, you need to know if the wait is CPU, disk, or locking — architecture tells you where to look.

See it live — copy this example

Run in SQL Server Management Studio (SSMS) or Azure Data Studio.

SELECT cpu_count, physical_memory_kb / 1024 AS MemoryMB
FROM sys.dm_os_sys_info;

SELECT name, type_desc, state_desc
FROM sys.databases
WHERE database_id > 4;

What happened?

• dm_os_sys_info shows CPU and memory the OS reports to SQL Server.
• The second query lists user databases and whether they are online — a quick health snapshot.

Practice next

1. Run the dm_os_sys_info query in SSMS.
2. Note CPU count and MemoryMB.
3. List user databases and confirm your database is ONLINE.
4. Query sys.dm_os_wait_stats TOP 10 by wait_time_ms.
5. Check SELECT name, recovery_model_desc FROM sys.databases;

Remember

Relational engine plans queries; storage engine moves pages. DMVs expose live instance health. User databases sit beside system databases.

Nightly report vs daytime OLTP

DataVerse OLTP shares the instance with a heavy report job.

Outcome: Team schedules reports off-peak after seeing CPU waits in DMVs.