cancel
Showing results for 
Search instead for 
Did you mean: 

How to Reduce backup window

Maddy777
Level 3

Current Backup solution runs on – NetBackup version 7.7.2. Latest release is known to be NetBackup 8.1.1.

Client has specified that the Current LTO5 Tape Backup solution takes longer backup window

Current Duration and Size

  1. Daily incremental backups
    Start at 5 PM Monday to approximately 3 Am the following morning
    2.28 TB (compressed) on average
  2. Weekly full backups
    Start at 5 PM on Friday, end approximately at 8 AM the following Monday morning
    15.36 TB (Compressed)

Is there anyway to find out why backup window is taking too long? Also, Please provide guidance as to how i can guide the client and what other questions and/ or data, I need to be asking or looking, besides:

  • Have you tested disk read speed on the client?
  • How fast can the client read its own data?
  • What kind of network link between server and client?
  • Have you tested network speed by ftp'ing some data from client to server?
  • Is Master/Media/Client to the same VLAN & to same switch?
  • What are you backing up on the client?
  • Have you enabled multiple data streams for data to be backed up?
  • Have you increased max jobs per client? (The default is 1)
  • Is Multiplexing enabled in Storage Unit and Policy schedules?

Any assistance is appreciated.

1 REPLY 1

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

I like your list of questions.

Each component in the backup path need to be investigated - most important is each client's ability to read from disk, followed by network interface and transfer rate to the media server.
Start with clients with lowest Kbytes/sec in Activity Monitor. 

When all of these components can get the data to the media server fast enough, look at size and number of data buffers on the media servers. 
Ensure bptm log folder exist on all media servers - buffer setup will be logged at the beginning of each job.
End-result of buffer usage will be logged at the end of each job.
Look for something like ' ... waited for full/empty buffers .... xxx times, delayed xxx times... '