PrepAway - Latest Free Exam Questions & Answers

Which recovery model should the database use?

You administer a Microsoft SQL Server 2012 instance that contains a financial database
hosted on a storage area network (SAN).
The financial database has the following characteristics:
• A data file of 2 terabytes is located on a dedicated LUN (drive D).
• A transaction log of 10 GB is located on a dedicated LUN (drive E).
• Drive D has 1 terabyte of free disk space.
• Drive E has 5 GB of free disk space.
The database is continually modified by users during business hours from Monday through
Friday between 09:00 hours and 17:00 hours. Five percent of the existing data is modified
each day.
The Finance department loads large CSV files into a number of tables each business day at
11:15 hours and 15:15 hours by using the BCP or BULK INSERT commands. Each data
load adds 3 GB of data to the database.
These data load operations must occur in the minimum amount of time.
A full database backup is performed every Sunday at 10:00 hours. Backup operations will be
performed every two hours (11:00, 13:00, 15:00, and 17:00) during business hours.
You need to ensure that the minimum amount of data is lost.
Which recovery model should the database use?

PrepAway - Latest Free Exam Questions & Answers

A.
FULL

B.
DBO_ONLY

C.
CONTINUE_AFTER_ERROR

D.
CHECKSUM

E.
NO_CHECKSUM

F.
SIMPLE

G.
Transaction log

H.
SKIP

I.
RESTART

J.
COPY_ONLY
K.
NORECOVERY
L.
BULK_LOGGED
M.
Differential
N.
STANDBY

Explanation:
http://msdn.microsoft.com/en-us/library/ms189275.aspx

20 Comments on “Which recovery model should the database use?

  1. JonBan says:

    Yet another tricky question… Full is indeed the only option with point in time recovery, allowing minimal data loss, but I would still answer BULK_LOGGED, because with FULL you can’t meet the “data load operations must occur in the minimum amount of time” requirement and, more importantly, the 10GB transaction log would certainly not be big enough.




    0



    0
  2. Slazenjer_m says:

    A very good reason to have FULL Recovery model…

    Scenario 1: “The database is continually modified by users during business hours from Monday through Friday between 09:00 hours and 17:00 hours. Five percent of the existing data is modified each day”

    Switch to BULK Recovery model to perform the data load & revert back (to FULL), to minimize data loss:

    Scenario 2: “The Finance department loads large CSV files into a number of tables each business day at 11:15 hours and 15:15 hours by using the BCP or BULK INSERT commands.”




    0



    0
  3. marcel says:

    I would say BULK_LOGGED
    because the say it must occur in the minimum
    These data load operations MUST occur in the minimum amount of time
    while FULL recovery gives you point in time recovery, still BULK_LOGGED gives you a full restore option when the DB fails.
    and point in time recovery is not a requirement.




    0



    0

Leave a Reply