 |
Page 1 of 1
|
| Author |
Message |
Philip Morrison
Guest
|
 Magnetic library size best practice.
What is the best size? We have a MR3000 attached with 30 TB capacity.. one big LUN or chop it up?
__._,_.___
Reply to sender ([email]pjmorr < at > intrasection.com?subject=Re%3A%20Magnetic%20library%20size%20best%20practice%2E[/email]) | Reply to group ([email]commvault < at > yahoogroups.com?subject=Re%3A%20Magnetic%20library%20size%20best%20practice%2E[/email]) | Reply via web post | Start a New Topic
Messages in this topic (1)
Recent Activity: Visit Your Group
Commvault Documentation here:
http://documentation.commvault.com/
Text-Only ([email]commvault-traditional < at > yahoogroups.com?subject=Change Delivery Format: Traditional[/email]), Daily Digest ([email]commvault-digest < at > yahoogroups.com?subject=Email Delivery: Digest[/email]) • Unsubscribe ([email]commvault-unsubscribe < at > yahoogroups.com?subject=Unsubscribe[/email]) • Terms of Use
.
[img]http://geo.yahoo.com/serv?s=97359714/grpId=7450851/grpspId=1707277520/msgId=4508/stime=1292474654/nc1=1/nc2=2/nc3=3[/img]
__,_._,___
|
| Wed Dec 15, 2010 8:44 pm |
|
 |
Mark Edwards
Guest
|
 Magnetic library size best practice.
chop it up.
On Thu, Dec 16, 2010 at 13:39, Philip Morrison <pjmorr < at > intrasection.com ([email]pjmorr < at > intrasection.com[/email])> wrote:
What is the best size? We have a MR3000 attached with 30 TB capacity.. one big LUN or chop it up?
--
regards,
Mark P. Edwards
__._,_.___
Reply to sender ([email]mark.p.edwards < at > gmail.com?subject=Re%3A%20%5Bcommvault%5D%20Magnetic%20library%20size%20best%20practice%2E[/email]) | Reply to group ([email]commvault < at > yahoogroups.com?subject=Re%3A%20%5Bcommvault%5D%20Magnetic%20library%20size%20best%20practice%2E[/email]) | Reply via web post | Start a New Topic
Messages in this topic (2)
Recent Activity: Visit Your Group
Commvault Documentation here:
http://documentation.commvault.com/
Text-Only ([email]commvault-traditional < at > yahoogroups.com?subject=Change Delivery Format: Traditional[/email]), Daily Digest ([email]commvault-digest < at > yahoogroups.com?subject=Email Delivery: Digest[/email]) • Unsubscribe ([email]commvault-unsubscribe < at > yahoogroups.com?subject=Unsubscribe[/email]) • Terms of Use
.
[img]http://geo.yahoo.com/serv?s=97359714/grpId=7450851/grpspId=1707277520/msgId=4509/stime=1292479956/nc1=1/nc2=2/nc3=3[/img]
__,_._,___
|
| Wed Dec 15, 2010 10:13 pm |
|
 |
Bill Willis
Guest
|
 Magnetic library size best practice.
The more pieces you can make it (but don't make the volumes too small), the more streams you can have writing to your 30 TB of disk.
Example:
1 30 TB lun, you have 1 mount point, so your jobs will run with fewer writers (or if you set a high number of writers, it's straining the disk).
30 1 TB luns, you set them to spill and fill (fill and spill will only write to 1 lun until it is full. Spill and Fill will write to random luns that have free space), and about 5 writers. You now have a lot of writers that are more evenly distributed over different areas of your disk, increasing your potential throughput.
1 TB may be small, it was just an easy example.
Philip Morrison <pjmorr < at > intrasection.com> 12/15/2010 6:39 PM >>>
What is the best size? We have a MR3000 attached with 30 TB capacity.. one big LUN or chop it up?
|
| Wed Dec 15, 2010 10:28 pm |
|
 |
Bart Donders
Guest
|
 Magnetic library size best practice.
I never use luns greater than 2 TB.
The explanation Bill provided is completely true.
--- In commvault < at > yahoogroups.com ([email]commvault%40yahoogroups.com[/email]), "Bill Willis" <bill.willis < at > ...> wrote:
The more pieces you can make it (but don't make the volumes too small), the more streams you can have writing to your 30 TB of disk.
Example:
1 30 TB lun, you have 1 mount point, so your jobs will run with fewer writers (or if you set a high number of writers, it's straining the disk).
30 1 TB luns, you set them to spill and fill (fill and spill will only write to 1 lun until it is full. Spill and Fill will write to random luns that have free space), and about 5 writers. You now have a lot of writers that are more evenly distributed over different areas of your disk, increasing your potential throughput.
1 TB may be small, it was just an easy example.
Philip Morrison <pjmorr < at > ...> 12/15/2010 6:39 PM >>>
What is the best size? We have a MR3000 attached with 30 TB capacity.. one big LUN or chop it up?
The information in this electronic mail communication (e-mail) contains confidential information which is the property of the sender and may be protected by the attorney-client privilege and/or attorney work product doctrine. It is intended solely for the addressee. Access to this e-mail by anyone else is unauthorized by the sender. If you are not the intended recipient, you are hereby notified that any disclosure, copying, or distribution of the contents of this e-mail transmission or the taking or omission of any action in reliance thereon or pursuant thereto, is prohibited, and may be unlawful. If you received this e-mail in error, please notify us immediately of your receipt of this message by e-mail and destroy this communication, any attachments, and all copies thereof.
Southwest Gas Corporation does not guarantee the privacy or security of information transmitted by facsimile or other unsecure electronic means. By choosing to send or receive confidential or personal identifying information via facsimile or unencrypted e-mail, you consent to accept any associated risk.
Thank you for your cooperation.
__._,_.___
Reply to sender ([email]commvault < at > sanessentials.nl?subject=Re%3A%20Magnetic%20library%20size%20best%20practice%2E[/email]) | Reply to group ([email]commvault < at > yahoogroups.com?subject=Re%3A%20Magnetic%20library%20size%20best%20practice%2E[/email]) | Reply via web post | Start a New Topic
Messages in this topic (4)
Recent Activity: Visit Your Group
Commvault Documentation here:
http://documentation.commvault.com/
Text-Only ([email]commvault-traditional < at > yahoogroups.com?subject=Change Delivery Format: Traditional[/email]), Daily Digest ([email]commvault-digest < at > yahoogroups.com?subject=Email Delivery: Digest[/email]) • Unsubscribe ([email]commvault-unsubscribe < at > yahoogroups.com?subject=Unsubscribe[/email]) • Terms of Use
.
[img]http://geo.yahoo.com/serv?s=97359714/grpId=7450851/grpspId=1707277520/msgId=4511/stime=1292500248/nc1=1/nc2=2/nc3=3[/img]
__,_._,___
|
| Thu Dec 16, 2010 3:50 am |
|
 |
Chris Dawson
Guest
|
 Magnetic library size best practice.
Pardon my ignorance but I don’t see how having (to continue using the simple example) 5 writers writing to one LUN would be worse than having 30 LUNs, 5 writers spread over the LUNs if the underlying disk layout hasn’t changed. If you have one large RAID 5 for example and carve that up in to a large number of LUNs then you are going to be having writes all over the place and the parity writing operations would have trouble keeping up. Or am I missing something here? Should we divide up 30TB into several RAIDs and distribute the LUNs across that?
Taken to the extreme, would it be worthwhile to forgo having disks in a RAID set and just use them as individual disks presented as LUNs? Would CV be able to deal gracefully with the much great likelihood of failure?
From: commvault < at > yahoogroups.com [mailto:commvault < at > yahoogroups.com] On Behalf Of Bart Donders
Sent: Thursday, 16 December 2010 10:51 PM
To: commvault < at > yahoogroups.com
Subject: [commvault] Re: Magnetic library size best practice.
I never use luns greater than 2 TB.
The explanation Bill provided is completely true.
--- In commvault < at > yahoogroups.com ([email]commvault%40yahoogroups.com[/email]), "Bill Willis" <bill.willis < at > ...> wrote:
The more pieces you can make it (but don't make the volumes too small), the more streams you can have writing to your 30 TB of disk.
Example:
1 30 TB lun, you have 1 mount point, so your jobs will run with fewer writers (or if you set a high number of writers, it's straining the disk).
30 1 TB luns, you set them to spill and fill (fill and spill will only write to 1 lun until it is full. Spill and Fill will write to random luns that have free space), and about 5 writers. You now have a lot of writers that are more evenly distributed over different areas of your disk, increasing your potential throughput.
1 TB may be small, it was just an easy example.
Philip Morrison <pjmorr < at > ...> 12/15/2010 6:39 PM >>>
What is the best size? We have a MR3000 attached with 30 TB capacity.. one big LUN or chop it up?
The information in this electronic mail communication (e-mail) contains confidential information which is the property of the sender and may be protected by the attorney-client privilege and/or attorney work product doctrine. It is intended solely for the addressee. Access to this e-mail by anyone else is unauthorized by the sender. If you are not the intended recipient, you are hereby notified that any disclosure, copying, or distribution of the contents of this e-mail transmission or the taking or omission of any action in reliance thereon or pursuant thereto, is prohibited, and may be unlawful. If you received this e-mail in error, please notify us immediately of your receipt of this message by e-mail and destroy this communication, any attachments, and all copies thereof.
Southwest Gas Corporation does not guarantee the privacy or security of information transmitted by facsimile or other unsecure electronic means. By choosing to send or receive confidential or personal identifying information via facsimile or unencrypted e-mail, you consent to accept any associated risk.
Thank you for your cooperation.
__._,_.___
Reply to sender ([email]chris.dawson < at > anu.edu.au?subject=RE%3A%20%5Bcommvault%5D%20Re%3A%20Magnetic%20library%20size%20best%20practice%2E[/email]) | Reply to group ([email]commvault < at > yahoogroups.com?subject=RE%3A%20%5Bcommvault%5D%20Re%3A%20Magnetic%20library%20size%20best%20practice%2E[/email]) | Reply via web post | Start a New Topic
Messages in this topic (5)
Recent Activity: Visit Your Group
Commvault Documentation here:
http://documentation.commvault.com/
MARKETPLACE
[url=http://global.ard.yahoo.com/SIG=15o9gi1mf/M=493064.13814537.14041040.10835568/D=groups/S=1707277520:MKP1/Y=YAHOO/EXP=1292825540/L=56ba718c-0bef-11e0-bb9a-278564293d59/B=qvqJEEwNO6g-/J=1292818340139229/K=QE9M7ESMVEWiuI1iinLFBg/A=6078812/R=0/SIG=114ae4ln1/*http://dogandcatanswers.yahoo.com/]Get great advice about dogs and cats. Visit the Dog & Cat Answers Center.[/url] if(window.yzq_d==null)window.yzq_d=new Object(); window.yzq_d['qvqJEEwNO6g-']='&U=13c3scqkc%2fN%3dqvqJEEwNO6g-%2fC%3d493064.13814537.14041040.10835568%2fD%3dMKP1%2fB%3d6078812%2fV%3d1'; [img]http://us.bc.yahoo.com/b?P=56ba718c-0bef-11e0-bb9a-278564293d59&T=1cgbkkkia%2fX%3d1292818340%2fE%3d1707277520%2fR%3dgroups%2fK%3d5%2fV%3d2.1%2fW%3dH%2fY%3dYAHOO%2fF%3d272543438%2fH%3dY29udGVudD0iR287V2lkZ2V0cztQb2RjYXN0cztNb2JpbGU7RmxpY2tyO0dyb3VwcztBdWN0aW9ucztTbWFsbF9CdXNpbmVzcztCcmllZmNhc2U7RmluYW5jZTsiIGRpc2FibGVzaHVmZmxpbmc9IjEiIHNlcnZlSWQ9IjU2YmE3MThjLTBiZWYtMTFlMC1iYjlhLTI3ODU2NDI5M2Q1OSIgc2l0ZUlkPSI0NDUyNTUxIiB0U3RtcD0iMTI5MjgxODM0MDEyNjU1OCIg%2fQ%3d-1%2fS%3d1%2fJ%3d3B5EC442&U=13c3scqkc%2fN%3dqvqJEEwNO6g-%2fC%3d493064.13814537.14041040.10835568%2fD%3dMKP1%2fB%3d6078812%2fV%3d1[/img]
[url=http://global.ard.yahoo.com/SIG=15odgaci3/M=493064.13983314.14041046.13298430/D=groups/S=1707277520:MKP1/Y=YAHOO/EXP=1292825540/L=56ba718c-0bef-11e0-bb9a-278564293d59/B=qfqJEEwNO6g-/J=1292818340139229/K=QE9M7ESMVEWiuI1iinLFBg/A=6060255/R=0/SIG=1194m4keh/*http://us.toolbar.yahoo.com/?.cpdl=grpj]Stay on top of your group activity without leaving the page you're on - Get the Yahoo! Toolbar now.[/url] if(window.yzq_d==null)window.yzq_d=new Object(); window.yzq_d['qfqJEEwNO6g-']='&U=13c6716nj%2fN%3dqfqJEEwNO6g-%2fC%3d493064.13983314.14041046.13298430%2fD%3dMKP1%2fB%3d6060255%2fV%3d1'; [img]http://us.bc.yahoo.com/b?P=56ba718c-0bef-11e0-bb9a-278564293d59&T=1ch3p5q8a%2fX%3d1292818340%2fE%3d1707277520%2fR%3dgroups%2fK%3d5%2fV%3d2.1%2fW%3dH%2fY%3dYAHOO%2fF%3d3199015319%2fH%3dY29udGVudD0iR287V2lkZ2V0cztQb2RjYXN0cztNb2JpbGU7RmxpY2tyO0dyb3VwcztBdWN0aW9ucztTbWFsbF9CdXNpbmVzcztCcmllZmNhc2U7RmluYW5jZTsiIGRpc2FibGVzaHVmZmxpbmc9IjEiIHNlcnZlSWQ9IjU2YmE3MThjLTBiZWYtMTFlMC1iYjlhLTI3ODU2NDI5M2Q1OSIgc2l0ZUlkPSI0NDUyNTUxIiB0U3RtcD0iMTI5MjgxODM0MDEyNjU1OCIg%2fQ%3d-1%2fS%3d1%2fJ%3d3B5EC442&U=13c6716nj%2fN%3dqfqJEEwNO6g-%2fC%3d493064.13983314.14041046.13298430%2fD%3dMKP1%2fB%3d6060255%2fV%3d1[/img]
[url=http://global.ard.yahoo.com/SIG=15olbjpgk/M=493064.14012770.13963757.13298430/D=groups/S=1707277520:MKP1/Y=YAHOO/EXP=1292825540/L=56ba718c-0bef-11e0-bb9a-278564293d59/B=q_qJEEwNO6g-/J=1292818340139229/K=QE9M7ESMVEWiuI1iinLFBg/A=6015306/R=0/SIG=11vlkvigg/*http://advision.webevents.yahoo.com/hobbiesandactivitieszone/]Hobbies & Activities Zone: Find others who share your passions! Explore new interests.[/url] if(window.yzq_d==null)window.yzq_d=new Object(); window.yzq_d['q_qJEEwNO6g-']='&U=13c1v7fin%2fN%3dq_qJEEwNO6g-%2fC%3d493064.14012770.13963757.13298430%2fD%3dMKP1%2fB%3d6015306%2fV%3d1'; [img]http://us.bc.yahoo.com/b?P=56ba718c-0bef-11e0-bb9a-278564293d59&T=1chsmoo45%2fX%3d1292818340%2fE%3d1707277520%2fR%3dgroups%2fK%3d5%2fV%3d2.1%2fW%3dH%2fY%3dYAHOO%2fF%3d1477662129%2fH%3dY29udGVudD0iR287V2lkZ2V0cztQb2RjYXN0cztNb2JpbGU7RmxpY2tyO0dyb3VwcztBdWN0aW9ucztTbWFsbF9CdXNpbmVzcztCcmllZmNhc2U7RmluYW5jZTsiIGRpc2FibGVzaHVmZmxpbmc9IjEiIHNlcnZlSWQ9IjU2YmE3MThjLTBiZWYtMTFlMC1iYjlhLTI3ODU2NDI5M2Q1OSIgc2l0ZUlkPSI0NDUyNTUxIiB0U3RtcD0iMTI5MjgxODM0MDEyNjU1OCIg%2fQ%3d-1%2fS%3d1%2fJ%3d3B5EC442&U=13c1v7fin%2fN%3dq_qJEEwNO6g-%2fC%3d493064.14012770.13963757.13298430%2fD%3dMKP1%2fB%3d6015306%2fV%3d1[/img]
Text-Only ([email]commvault-traditional < at > yahoogroups.com?subject=Change Delivery Format: Traditional[/email]), Daily Digest ([email]commvault-digest < at > yahoogroups.com?subject=Email Delivery: Digest[/email]) • Unsubscribe ([email]commvault-unsubscribe < at > yahoogroups.com?subject=Unsubscribe[/email]) • Terms of Use
.
[img]http://geo.yahoo.com/serv?s=97359714/grpId=7450851/grpspId=1707277520/msgId=4512/stime=1292818339/nc1=5733769/nc2=4025304/nc3=4507179[/img]
__,_._,___
|
| Sun Dec 19, 2010 8:12 pm |
|
 |
Luke
Guest
|
 Magnetic library size best practice.
I'm a bit shaky on the storage side but AFAIK having 30TB in one large RAID 5 would not only result in poor performance but a rebuild would take more time to complete than would be acceptable...
Considering CommVault leaves I/O up to the FIle System (it's not going to treat the disk as RAW and do writes to different platters - its just going to write to a set of files), it would make more sense to split the 30TB into multiple RAID 5 sets to ensure that you get an (mostly) even distribution of I/O across the platters.
Alternative RAID options might be worth considering depending on what your storage array offers and the performance speed off the back of those options .vs. any potential risk for disruption to your disk library/magnetic library.
On 20/12/2010, at 11:26 AM, Chris Dawson wrote:
Pardon my ignorance but I don’t see how having (to continue using the simple example) 5 writers writing to one LUN would be worse than having 30 LUNs, 5 writers spread over the LUNs if the underlying disk layout hasn’t changed. If you have one large RAID 5 for example and carve that up in to a large number of LUNs then you are going to be having writes all over the place and the parity writing operations would have trouble keeping up. Or am I missing something here? Should we divide up 30TB into several RAIDs and distribute the LUNs across that?
Taken to the extreme, would it be worthwhile to forgo having disks in a RAID set and just use them as individual disks presented as LUNs? Would CV be able to deal gracefully with the much great likelihood of failure?
From: commvault < at > yahoogroups.com ([email]commvault < at > yahoogroups.com[/email]) [mailto:commvault < at > yahoogroups.com] On Behalf Of Bart Donders
Sent: Thursday, 16 December 2010 10:51 PM
To: commvault < at > yahoogroups.com ([email]commvault < at > yahoogroups.com[/email])
Subject: [commvault] Re: Magnetic library size best practice.
I never use luns greater than 2 TB.
The explanation Bill provided is completely true.
--- In commvault < at > yahoogroups.com ([email]commvault%40yahoogroups.com[/email]), "Bill Willis" <bill.willis < at > ...> wrote:
The more pieces you can make it (but don't make the volumes too small), the more streams you can have writing to your 30 TB of disk.
Example:
1 30 TB lun, you have 1 mount point, so your jobs will run with fewer writers (or if you set a high number of writers, it's straining the disk).
30 1 TB luns, you set them to spill and fill (fill and spill will only write to 1 lun until it is full. Spill and Fill will write to random luns that have free space), and about 5 writers. You now have a lot of writers that are more evenly distributed over different areas of your disk, increasing your potential throughput.
1 TB may be small, it was just an easy example.
Philip Morrison <pjmorr < at > ...> 12/15/2010 6:39 PM >>>
What is the best size? We have a MR3000 attached with 30 TB capacity.. one big LUN or chop it up?
The information in this electronic mail communication (e-mail) contains confidential information which is the property of the sender and may be protected by the attorney-client privilege and/or attorney work product doctrine. It is intended solely for the addressee. Access to this e-mail by anyone else is unauthorized by the sender. If you are not the intended recipient, you are hereby notified that any disclosure, copying, or distribution of the contents of this e-mail transmission or the taking or omission of any action in reliance thereon or pursuant thereto, is prohibited, and may be unlawful. If you received this e-mail in error, please notify us immediately of your receipt of this message by e-mail and destroy this communication, any attachments, and all copies thereof.
Southwest Gas Corporation does not guarantee the privacy or security of information transmitted by facsimile or other unsecure electronic means. By choosing to send or receive confidential or personal identifying information via facsimile or unencrypted e-mail, you consent to accept any associated risk.
Thank you for your cooperation.
__._,_.___
Reply to sender ([email]kaneda.san.sec < at > gmail.com?subject=Re%3A%20%5Bcommvault%5D%20Re%3A%20Magnetic%20library%20size%20best%20practice%2E[/email]) | Reply to group ([email]commvault < at > yahoogroups.com?subject=Re%3A%20%5Bcommvault%5D%20Re%3A%20Magnetic%20library%20size%20best%20practice%2E[/email]) | Reply via web post | Start a New Topic
Messages in this topic (6)
Recent Activity: Visit Your Group
Commvault Documentation here:
http://documentation.commvault.com/
MARKETPLACE
[url=http://global.ard.yahoo.com/SIG=15om2lml2/M=493064.13983314.14041046.13298430/D=groups/S=1707277520:MKP1/Y=YAHOO/EXP=1292826303/L=1d756754-0bf1-11e0-bcf6-e31dda93f70d/B=P4lQCGKJiSM-/J=1292819103025397/K=kfHi8sKadiEj_Y8y_UCTmw/A=6060255/R=0/SIG=1194m4keh/*http://us.toolbar.yahoo.com/?.cpdl=grpj]Stay on top of your group activity without leaving the page you're on - Get the Yahoo! Toolbar now.[/url] if(window.yzq_d==null)window.yzq_d=new Object(); window.yzq_d['P4lQCGKJiSM-']='&U=13cov0d56%2fN%3dP4lQCGKJiSM-%2fC%3d493064.13983314.14041046.13298430%2fD%3dMKP1%2fB%3d6060255%2fV%3d1'; [img]http://us.bc.yahoo.com/b?P=1d756754-0bf1-11e0-bcf6-e31dda93f70d&T=1ch4ia2gu%2fX%3d1292819103%2fE%3d1707277520%2fR%3dgroups%2fK%3d5%2fV%3d2.1%2fW%3dH%2fY%3dYAHOO%2fF%3d3436385266%2fH%3dY29udGVudD0iR287V2lkZ2V0cztQb2RjYXN0cztNb2JpbGU7RmxpY2tyO0dyb3VwcztBdWN0aW9ucztTbWFsbF9CdXNpbmVzcztCcmllZmNhc2U7RmluYW5jZTsiIGRpc2FibGVzaHVmZmxpbmc9IjEiIHNlcnZlSWQ9IjFkNzU2NzU0LTBiZjEtMTFlMC1iY2Y2LWUzMWRkYTkzZjcwZCIgc2l0ZUlkPSI0NDUyNTUxIiB0U3RtcD0iMTI5MjgxOTEwMzAzNzQxNyIg%2fQ%3d-1%2fS%3d1%2fJ%3d27228962&U=13cov0d56%2fN%3dP4lQCGKJiSM-%2fC%3d493064.13983314.14041046.13298430%2fD%3dMKP1%2fB%3d6060255%2fV%3d1[/img]
[url=http://global.ard.yahoo.com/SIG=15op66o85/M=493064.14012770.13963757.13298430/D=groups/S=1707277520:MKP1/Y=YAHOO/EXP=1292826303/L=1d756754-0bf1-11e0-bcf6-e31dda93f70d/B=QYlQCGKJiSM-/J=1292819103025397/K=kfHi8sKadiEj_Y8y_UCTmw/A=6015306/R=0/SIG=11vlkvigg/*http://advision.webevents.yahoo.com/hobbiesandactivitieszone/]Hobbies & Activities Zone: Find others who share your passions! Explore new interests.[/url] if(window.yzq_d==null)window.yzq_d=new Object(); window.yzq_d['QYlQCGKJiSM-']='&U=13c2jrdm0%2fN%3dQYlQCGKJiSM-%2fC%3d493064.14012770.13963757.13298430%2fD%3dMKP1%2fB%3d6015306%2fV%3d1'; [img]http://us.bc.yahoo.com/b?P=1d756754-0bf1-11e0-bcf6-e31dda93f70d&T=1ch836l1h%2fX%3d1292819103%2fE%3d1707277520%2fR%3dgroups%2fK%3d5%2fV%3d2.1%2fW%3dH%2fY%3dYAHOO%2fF%3d3891508127%2fH%3dY29udGVudD0iR287V2lkZ2V0cztQb2RjYXN0cztNb2JpbGU7RmxpY2tyO0dyb3VwcztBdWN0aW9ucztTbWFsbF9CdXNpbmVzcztCcmllZmNhc2U7RmluYW5jZTsiIGRpc2FibGVzaHVmZmxpbmc9IjEiIHNlcnZlSWQ9IjFkNzU2NzU0LTBiZjEtMTFlMC1iY2Y2LWUzMWRkYTkzZjcwZCIgc2l0ZUlkPSI0NDUyNTUxIiB0U3RtcD0iMTI5MjgxOTEwMzAzNzQxNyIg%2fQ%3d-1%2fS%3d1%2fJ%3d27228962&U=13c2jrdm0%2fN%3dQYlQCGKJiSM-%2fC%3d493064.14012770.13963757.13298430%2fD%3dMKP1%2fB%3d6015306%2fV%3d1[/img]
[url=http://global.ard.yahoo.com/SIG=15oolqu80/M=493064.13814537.14041040.10835568/D=groups/S=1707277520:MKP1/Y=YAHOO/EXP=1292826303/L=1d756754-0bf1-11e0-bcf6-e31dda93f70d/B=QIlQCGKJiSM-/J=1292819103025397/K=kfHi8sKadiEj_Y8y_UCTmw/A=6260316/R=0/SIG=11bc8n8gt/*http://custom.yahoo.com/fibromyalgiazone/]Find useful articles and helpful tips on living with Fibromyalgia. Visit the Fibromyalgia Zone today![/url] if(window.yzq_d==null)window.yzq_d=new Object(); window.yzq_d['QIlQCGKJiSM-']='&U=13cpiolr2%2fN%3dQIlQCGKJiSM-%2fC%3d493064.13814537.14041040.10835568%2fD%3dMKP1%2fB%3d6260316%2fV%3d1'; [img]http://us.bc.yahoo.com/b?P=1d756754-0bf1-11e0-bcf6-e31dda93f70d&T=1chr3c3sl%2fX%3d1292819103%2fE%3d1707277520%2fR%3dgroups%2fK%3d5%2fV%3d2.1%2fW%3dH%2fY%3dYAHOO%2fF%3d1578550292%2fH%3dY29udGVudD0iR287V2lkZ2V0cztQb2RjYXN0cztNb2JpbGU7RmxpY2tyO0dyb3VwcztBdWN0aW9ucztTbWFsbF9CdXNpbmVzcztCcmllZmNhc2U7RmluYW5jZTsiIGRpc2FibGVzaHVmZmxpbmc9IjEiIHNlcnZlSWQ9IjFkNzU2NzU0LTBiZjEtMTFlMC1iY2Y2LWUzMWRkYTkzZjcwZCIgc2l0ZUlkPSI0NDUyNTUxIiB0U3RtcD0iMTI5MjgxOTEwMzAzNzQxNyIg%2fQ%3d-1%2fS%3d1%2fJ%3d27228962&U=13cpiolr2%2fN%3dQIlQCGKJiSM-%2fC%3d493064.13814537.14041040.10835568%2fD%3dMKP1%2fB%3d6260316%2fV%3d1[/img]
Text-Only ([email]commvault-traditional < at > yahoogroups.com?subject=Change Delivery Format: Traditional[/email]), Daily Digest ([email]commvault-digest < at > yahoogroups.com?subject=Email Delivery: Digest[/email]) • Unsubscribe ([email]commvault-unsubscribe < at > yahoogroups.com?subject=Unsubscribe[/email]) • Terms of Use
.
[img]http://geo.yahoo.com/serv?s=97359714/grpId=7450851/grpspId=1707277520/msgId=4513/stime=1292819102/nc1=5733763/nc2=4025338/nc3=4507179[/img]
__,_._,___
|
| Sun Dec 19, 2010 8:25 pm |
|
 |
Mark Edwards
Guest
|
 Magnetic library size best practice.
You're getting into the nuts and bolts of performance tuning here. Bascially there are a number of bottlenecks in any storage infrastructure and the trick is to find the most effective way of minimising the effect of the smallest bottleneck in the chain.
From a storage perspective, you would want to lay out your storage LUNs in such a way as to maximise throughput and redundancy across SAN controllers, storage spindle paths and cache.
At the Media Agent end, you're looking to run as many writers as you possibly can before saturating the controller cache and the SAN path.
All of the components in the storage chain have different performance tuneable parameters and the best solution is going to differ between different designs of storage backend.
If you want to consider using lone spindles as storage for a commvault maglib, then my recommendation would be to only use these as spool maglibs. No application will handle loss of data from a storage infrastructure like this very gracefully indeed. If you want to keep your data available from disk then you will need to implement some sort of redundant layout.
On Mon, Dec 20, 2010 at 11:26, Chris Dawson <chris.dawson < at > anu.edu.au ([email]chris.dawson < at > anu.edu.au[/email])> wrote:
Pardon my ignorance but I don’t see how having (to continue using the simple example) 5 writers writing to one LUN would be worse than having 30 LUNs, 5 writers spread over the LUNs if the underlying disk layout hasn’t changed. If you have one large RAID 5 for example and carve that up in to a large number of LUNs then you are going to be having writes all over the place and the parity writing operations would have trouble keeping up. Or am I missing something here? Should we divide up 30TB into several RAIDs and distribute the LUNs across that?
Taken to the extreme, would it be worthwhile to forgo having disks in a RAID set and just use them as individual disks presented as LUNs? Would CV be able to deal gracefully with the much great likelihood of failure?
From: commvault < at > yahoogroups.com ([email]commvault < at > yahoogroups.com[/email]) [mailto:commvault < at > yahoogroups.com ([email]commvault < at > yahoogroups.com[/email])] On Behalf Of Bart Donders
Sent: Thursday, 16 December 2010 10:51 PM
To: commvault < at > yahoogroups.com ([email]commvault < at > yahoogroups.com[/email])
Subject: [commvault] Re: Magnetic library size best practice.
I never use luns greater than 2 TB.
The explanation Bill provided is completely true.
--- In commvault < at > yahoogroups.com ([email]commvault%40yahoogroups.com[/email]), "Bill Willis" <bill.willis < at > ...> wrote:
The more pieces you can make it (but don't make the volumes too small), the more streams you can have writing to your 30 TB of disk.
Example:
1 30 TB lun, you have 1 mount point, so your jobs will run with fewer writers (or if you set a high number of writers, it's straining the disk).
30 1 TB luns, you set them to spill and fill (fill and spill will only write to 1 lun until it is full. Spill and Fill will write to random luns that have free space), and about 5 writers. You now have a lot of writers that are more evenly distributed over different areas of your disk, increasing your potential throughput.
1 TB may be small, it was just an easy example.
Philip Morrison <pjmorr < at > ...> 12/15/2010 6:39 PM >>>
What is the best size? We have a MR3000 attached with 30 TB capacity.. one big LUN or chop it up?
The information in this electronic mail communication (e-mail) contains confidential information which is the property of the sender and may be protected by the attorney-client privilege and/or attorney work product doctrine. It is intended solely for the addressee. Access to this e-mail by anyone else is unauthorized by the sender. If you are not the intended recipient, you are hereby notified that any disclosure, copying, or distribution of the contents of this e-mail transmission or the taking or omission of any action in reliance thereon or pursuant thereto, is prohibited, and may be unlawful. If you received this e-mail in error, please notify us immediately of your receipt of this message by e-mail and destroy this communication, any attachments, and all copies thereof.
Southwest Gas Corporation does not guarantee the privacy or security of information transmitted by facsimile or other unsecure electronic means. By choosing to send or receive confidential or personal identifying information via facsimile or unencrypted e-mail, you consent to accept any associated risk.
Thank you for your cooperation.
--
regards,
Mark P. Edwards
__._,_.___
Reply to sender ([email]mark.p.edwards < at > gmail.com?subject=Re%3A%20%5Bcommvault%5D%20Re%3A%20Magnetic%20library%20size%20best%20practice%2E[/email]) | Reply to group ([email]commvault < at > yahoogroups.com?subject=Re%3A%20%5Bcommvault%5D%20Re%3A%20Magnetic%20library%20size%20best%20practice%2E[/email]) | Reply via web post | Start a New Topic
Messages in this topic (7)
Recent Activity: Visit Your Group
Commvault Documentation here:
http://documentation.commvault.com/
Text-Only ([email]commvault-traditional < at > yahoogroups.com?subject=Change Delivery Format: Traditional[/email]), Daily Digest ([email]commvault-digest < at > yahoogroups.com?subject=Email Delivery: Digest[/email]) • Unsubscribe ([email]commvault-unsubscribe < at > yahoogroups.com?subject=Unsubscribe[/email]) • Terms of Use
.
[img]http://geo.yahoo.com/serv?s=97359714/grpId=7450851/grpspId=1707277520/msgId=4514/stime=1292830053/nc1=1/nc2=2/nc3=3[/img]
__,_._,___
|
| Sun Dec 19, 2010 11:27 pm |
|
 |
Bill Willis
Guest
|
 Magnetic library size best practice.
Back when I upgraded to CV 7 a couple of years back, this is what my environment looked like.
I had a NexSAN Satabeast with 16 usable TB of space (20 TB raw). I mounted it as a single 16 TB disk onto the media agent. My read/write speed on the disk was somewhere around 30-35 Megabytes per second.
The Commvault SE who was helping us with our upgrade suggested I turn it into 16 1 TB volumes instead. That one change made my disk read/write speed (measured with iometer) jump up to over 200 MB/Sec. My Aux copies over fiber increased by a huge margin.
Mark Edwards <mark.p.edwards < at > gmail.com> 12/19/2010 11:27 PM >>>
You're getting into the nuts and bolts of performance tuning here. Bascially there are a number of bottlenecks in any storage infrastructure and the trick is to find the most effective way of minimising the effect of the smallest bottleneck in the chain.
From a storage perspective, you would want to lay out your storage LUNs in such a way as to maximise throughput and redundancy across SAN controllers, storage spindle paths and cache.
At the Media Agent end, you're looking to run as many writers as you possibly can before saturating the controller cache and the SAN path.
All of the components in the storage chain have different performance tuneable parameters and the best solution is going to differ between different designs of storage backend.
If you want to consider using lone spindles as storage for a commvault maglib, then my recommendation would be to only use these as spool maglibs. No application will handle loss of data from a storage infrastructure like this very gracefully indeed. If you want to keep your data available from disk then you will need to implement some sort of redundant layout.
On Mon, Dec 20, 2010 at 11:26, Chris Dawson <chris.dawson < at > anu.edu.au ([email]chris.dawson < at > anu.edu.au[/email])> wrote:
Pardon my ignorance but I don't see how having (to continue using the simple example) 5 writers writing to one LUN would be worse than having 30 LUNs, 5 writers spread over the LUNs if the underlying disk layout hasn't changed. If you have one large RAID 5 for example and carve that up in to a large number of LUNs then you are going to be having writes all over the place and the parity writing operations would have trouble keeping up. Or am I missing something here? Should we divide up 30TB into several RAIDs and distribute the LUNs across that?
Taken to the extreme, would it be worthwhile to forgo having disks in a RAID set and just use them as individual disks presented as LUNs? Would CV be able to deal gracefully with the much great likelihood of failure?
From: commvault < at > yahoogroups.com ([email]commvault < at > yahoogroups.com[/email]) [mailto:commvault < at > yahoogroups.com ([email]commvault < at > yahoogroups.com[/email])] On Behalf Of Bart Donders
Sent: Thursday, 16 December 2010 10:51 PM
To: commvault < at > yahoogroups.com ([email]commvault < at > yahoogroups.com[/email])
Subject: [commvault] Re: Magnetic library size best practice.
I never use luns greater than 2 TB.
The explanation Bill provided is completely true.
--- In commvault < at > yahoogroups.com ([email]commvault%40yahoogroups.com[/email]), "Bill Willis" <bill.willis < at > ...> wrote:
The more pieces you can make it (but don't make the volumes too small), the more streams you can have writing to your 30 TB of disk.
Example:
1 30 TB lun, you have 1 mount point, so your jobs will run with fewer writers (or if you set a high number of writers, it's straining the disk).
30 1 TB luns, you set them to spill and fill (fill and spill will only write to 1 lun until it is full. Spill and Fill will write to random luns that have free space), and about 5 writers. You now have a lot of writers that are more evenly distributed over different areas of your disk, increasing your potential throughput.
1 TB may be small, it was just an easy example.
Philip Morrison <pjmorr < at > ...> 12/15/2010 6:39 PM >>>
What is the best size? We have a MR3000 attached with 30 TB capacity.. one big LUN or chop it up?
The information in this electronic mail communication (e-mail) contains confidential information which is the property of the sender and may be protected by the attorney-client privilege and/or attorney work product doctrine. It is intended solely for the addressee. Access to this e-mail by anyone else is unauthorized by the sender. If you are not the intended recipient, you are hereby notified that any disclosure, copying, or distribution of the contents of this e-mail transmission or the taking or omission of any action in reliance thereon or pursuant thereto, is prohibited, and may be unlawful. If you received this e-mail in error, please notify us immediately of your receipt of this message by e-mail and destroy this communication, any attachments, and all copies thereof.
Southwest Gas Corporation does not guarantee the privacy or security of information transmitted by facsimile or other unsecure electronic means. By choosing to send or receive confidential or personal identifying information via facsimile or unencrypted e-mail, you consent to accept any associated risk.
Thank you for your cooperation.
--
regards,
Mark P. Edwards
|
| Mon Dec 20, 2010 8:20 am |
|
 |
|
|
The time now is Fri May 25, 2012 7:42 am | All times are GMT - 8 Hours
|
Page 1 of 1
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
|
|