cancel
Showing results forΒ 
Search instead forΒ 
Did you mean:Β 

accelerator, change journal, change time, archive bit

stucci
Level 6

Hi all,

I need to backup a big size data. The backup is already running and I search to improve the performance.

Some information about my environment:

NBU master Linux RH NBU 7.6.1.2

NBU media Windows 2012 Standard NBU 7.6.1.2

The media server write directly to the HP StoreOnce. The data to save are *.bak files each of 500GB size. I tried some way to increase the performance, in first time I knew that the StoreOnce does not support synthetic backup, so I used the accelerator. I write below the job details with accelerator Enabled, when the media can write beetwin 80 and 100 MB/Sec:

29-mar-2016 11.32.26 - Info bpbrm (pid=156) pdwbck-bk860 is the host to backup data from
29-mar-2016 11.32.26 - Info bpbrm (pid=156) reading file list for client
29-mar-2016 11.32.26 - Info bpbrm (pid=156) accelerator enabled
29-mar-2016 11.32.26 - Info bpbrm (pid=156) starting bpbkar32 on client
29-mar-2016 11.32.27 - Info bpbkar32 (pid=9500) Backup started
29-mar-2016 11.32.31 - Info bpbkar32 (pid=9500) change time comparison:<enabled>
29-mar-2016 11.32.31 - Info bpbkar32 (pid=9500) accelerator enabled backup, archive bit processing:<disabled>
29-mar-2016 11.32.31 - Info bptm (pid=13452) start
29-mar-2016 11.32.32 - Info bpbkar32 (pid=9500) will attempt to use change journal data for <E:\backups\LIQ70>
29-mar-2016 11.32.36 - Info bptm (pid=13452) using 262144 data buffer size
29-mar-2016 11.32.36 - Info bptm (pid=13452) setting receive network buffer to 1049600 bytes
29-mar-2016 11.32.36 - Info bptm (pid=13452) using 512 data buffers
29-mar-2016 11.32.36 - Info bptm (pid=13452) start backup
29-mar-2016 11.33.13 - Info bpbkar32 (pid=9500) not using change journal data for <E:\backups\LIQ70>: unable to validate change journal usage <reason=previous backups do not contain valid USNs>
29-mar-2016 11.33.13 - Info bpbkar32 (pid=9500) not using change journal data for <E:\backups\LIQ70>: forcing rescan, each file will be read in order to validate checksums
29-mar-2016 11.43.19 - Info nbjm (pid=16069) starting backup job (jobid=23665) for client pdwbck-bk860, policy PDWH_db_LIQ70, schedule FULL
29-mar-2016 11.43.19 - Info nbjm (pid=16069) requesting STANDARD_RESOURCE resources from RB for backup job (jobid=23665, request id:{AB85119E-F592-11E5-9ECD-28242E286BD1})
29-mar-2016 11.43.19 - requesting resource STU_ccmso01-bk860_PDWH
29-mar-2016 11.43.19 - requesting resource nbumaster02.NBU_CLIENT.MAXJOBS.pdwbck-bk860
29-mar-2016 11.43.19 - requesting resource nbumaster02.NBU_POLICY.MAXJOBS.PDWH_db_LIQ70
29-mar-2016 11.43.19 - granted resource  nbumaster02.NBU_CLIENT.MAXJOBS.pdwbck-bk860
29-mar-2016 11.43.19 - granted resource  nbumaster02.NBU_POLICY.MAXJOBS.PDWH_db_LIQ70
29-mar-2016 11.43.19 - granted resource  MediaID=@aaabl;DiskVolume=CCMSO01_PDWH;DiskPool=DP_ccmso01-bk860_PDWH;Path=CCMSO01_PDWH;StorageServer=ccmso01-bk860.sede.corp.sanpaoloimi.com;MediaServer=pdwbck
29-mar-2016 11.43.19 - granted resource  STU_ccmso01-bk860_PDWH
29-mar-2016 11.43.19 - estimated 0 kbytes needed
29-mar-2016 11.43.19 - Info nbjm (pid=16069) started backup (backupid=pdwbck-bk860_1459244599) job for client pdwbck-bk860, policy PDWH_db_LIQ70, schedule FULL on storage unit STU_ccmso01-bk860_PDWH
29-mar-2016 11.43.20 - started process bpbrm (pid=156)
29-mar-2016 11.43.20 - connecting
29-mar-2016 11.43.20 - connected; connect time: 0:00:00
29-mar-2016 11.43.39 - begin writing

I used this configuration for some Month but the performance are not very good and I have continuosly problem, often the backup fail. Today I read some tech note and I tried to disable the accelerator, and I was impressed to see the media to write around 180 MB/Sec. Below the job details:

29-mar-2016 14.20.48 - Info bpbrm (pid=11388) pdwbck-bk860 is the host to backup data from
29-mar-2016 14.20.49 - Info bpbrm (pid=11388) reading file list for client
29-mar-2016 14.20.49 - Info bpbrm (pid=11388) starting bpbkar32 on client
29-mar-2016 14.20.49 - Info bpbkar32 (pid=6312) Backup started
29-mar-2016 14.20.49 - Info bpbkar32 (pid=6312) change time comparison:<disabled>
29-mar-2016 14.20.49 - Info bpbkar32 (pid=6312) archive bit processing:<enabled>
29-mar-2016 14.20.49 - Info bptm (pid=14656) start
29-mar-2016 14.20.50 - Info bpbkar32 (pid=6312) not using change journal data for <E:\backups\LIQ70>: not configured for use
29-mar-2016 14.20.54 - Info bptm (pid=14656) using 262144 data buffer size
29-mar-2016 14.20.54 - Info bptm (pid=14656) setting receive network buffer to 1049600 bytes
29-mar-2016 14.20.54 - Info bptm (pid=14656) using 512 data buffers
29-mar-2016 14.20.54 - Info bptm (pid=14656) start backup
29-mar-2016 14.31.41 - Info nbjm (pid=16069) starting backup job (jobid=23837) for client pdwbck-bk860, policy PDWH_db_LIQ70, schedule FULL
29-mar-2016 14.31.41 - Info nbjm (pid=16069) requesting STANDARD_RESOURCE resources from RB for backup job (jobid=23837, request id:{30B13A52-F5AA-11E5-B957-5CC1C1356DC8})
29-mar-2016 14.31.41 - requesting resource STU_ccmso01-bk860_PDWH
29-mar-2016 14.31.41 - requesting resource nbumaster02.NBU_CLIENT.MAXJOBS.pdwbck-bk860
29-mar-2016 14.31.41 - requesting resource nbumaster02.NBU_POLICY.MAXJOBS.PDWH_db_LIQ70
29-mar-2016 14.31.41 - granted resource  nbumaster02.NBU_CLIENT.MAXJOBS.pdwbck-bk860
29-mar-2016 14.31.41 - granted resource  nbumaster02.NBU_POLICY.MAXJOBS.PDWH_db_LIQ70
29-mar-2016 14.31.41 - granted resource  MediaID=@aaabl;DiskVolume=CCMSO01_PDWH;DiskPool=DP_ccmso01-bk860_PDWH;Path=CCMSO01_PDWH;StorageServer=ccmso01-bk860.sede.corp.sanpaoloimi.com;MediaServer=pdwbck
29-mar-2016 14.31.41 - granted resource  STU_ccmso01-bk860_PDWH
29-mar-2016 14.31.42 - estimated 0 kbytes needed
29-mar-2016 14.31.42 - Info nbjm (pid=16069) started backup (backupid=pdwbck-bk860_1459254702) job for client pdwbck-bk860, policy PDWH_db_LIQ70, schedule FULL on storage unit STU_ccmso01-bk860_PDWH
29-mar-2016 14.31.42 - started process bpbrm (pid=11388)
29-mar-2016 14.31.43 - connecting
29-mar-2016 14.31.43 - connected; connect time: 0:00:00
29-mar-2016 14.31.52 - begin writing

So I started to analize about:

  • change time comparison
  • archive bit processing
  • change journal

Now, my questions, by default on Media/Client "archive bit processing" is Enabled, but I saw that when I use the accelerator:

archive bit processing = disabled; change time comparison = enabled

It is normal this behavior ?

Second question, do you think that the "change journal" can help me to write more fast the data ?

 

thank you very much,

regards 

 

3 REPLIES 3

Backup_All
Level 2

I don't think the Accelerator will increase your backup speed; it will reduce the amount of data you are backing up.

Deeps
Level 6

For accelerator backups , archive bit processing = disabled; change time comparison = enabled is normal.

In a traditional incremental backup without the use of Accelerator, the Nbu client would have to check evey single file on the client's volume to see which ones have been changed since the last backup, via the archive-bit or time stamp. Say the client has a total of 100 files, then during an incremental backup, all 100 files will have to be enumerated and checked. This takes alot of time.

With Accelerator, the changed blocks since the last backup are tracked. So, in the next incremental backup, only these changed blocks will be sent. Therefor change time comparison is enabled.

 

And as mentioend by Backup_All, Accelerator backup  helps in sending less data ( only changed blocks) over network for backup and effectively in shorter backup windows.

 

DeepS

stucci
Level 6

I tried to move the backup on DataDomain and the accelerator work fine, and to wite 14 TB of data employ 2 hour.

I hate StoreOnce :)

regards