Post by thread:[TECH] Seemingly slow backup over network

I have run a backup over a network. It has transferred approximately 19 GBytes in a little over 5 hours on a 100BaseT network. [[ 19.29 GByte & ( 5:14 hours ==> 18.84K seconds ) ==> 1.204 MByte/sec or ~ 12Mbits/sec ]] Is this to be expected or is the network being under-utilized ( there is no other traffic ) gu

On 25 Feb 2012 at 19:54, YES NOPE9 wrote: > I have run a backup over a network. It has transferred approximately > 19 GBytes in a little over 5 hours on a 100BaseT network. [[ 19.29 > GByte & ( 5:14 hours ==> 18.84K seconds ) ==> 1.204 MByte/sec or > ~ 12Mbits/sec ]] Is this to be expected or is the network being > under-utilized ( there is no other traffic ) gus Yes, it does seem slow. Many, many variables to consider of course, but by comparison I have a Gbit (1000BaseT) network and clocked it at something like 35MByte/sec on a large transfer. -- Brent Brown, Electronic Design Solutions 16 English Street, St Andrews, Hamilton 3200, New Zealand Ph: +64 7 849 0069 Fax: +64 7 849 0071 Cell: +64 27 433 4069 eMail: spam_OUTbrent.brownTakeThisOuTclear.net.nz

Wow 29 times faster {Quote hidden}> On Feb 25, 2012, at 9:11 PM, Brent Brown wrote: > > On 25 Feb 2012 at 19:54, YES NOPE9 wrote: > >> I have run a backup over a network. It has transferred approximately >> 19 GBytes in a little over 5 hours on a 100BaseT network. [[ 19.29 >> GByte & ( 5:14 hours ==> 18.84K seconds ) ==> 1.204 MByte/sec or >> ~ 12Mbits/sec ]] Is this to be expected or is the network being >> under-utilized ( there is no other traffic ) gus > > Yes, it does seem slow. Many, many variables to consider of course, but by > comparison I have a Gbit (1000BaseT) network and clocked it at something like > 35MByte/sec on a large transfer. > > -- > Brent Brown, Electronic Design Solutions > 16 English Street, St Andrews, > Hamilton 3200, New Zealand > Ph: +64 7 849 0069 > Fax: +64 7 849 0071 > Cell: +64 27 433 4069 > eMail: .....brent.brownKILLspam@spam@clear.net.nz >

>> On 25 Feb 2012 at 19:54, YES NOPE9 wrote: >> >>> I have run a backup over a network. It has transferred approximately >>> 19 GBytes in a little over 5 hours on a 100BaseT network. [[ 19.29 >>> GByte & ( 5:14 hours ==> 18.84K seconds ) ==> 1.204 MByte/sec or >>> ~ 12Mbits/sec ]] Is this to be expected or is the network being >>> under-utilized ( there is no other traffic ) gus 1.2 to 1.3 MBps seems low for 100Mbps Ethernet signalling rate. If window sizes, network adapters, intermediate Ethernet switches, duplex settings, system buses, etc are all set up correctly, you should see 5 to 9 MBps end to end transfer rates. >> On Feb 25, 2012, at 9:11 PM, Brent Brown wrote: >> >> Yes, it does seem slow. Many, many variables to consider of course, >> but by comparison I have a Gbit (1000BaseT) network and clocked it >> at something like 35MByte/sec on a large transfer. > > On 25 Feb 2012 at 20:59-PST, YES NOPE9 wrote: > > Wow 29 times faster Yes, but consider that 10X of that improvement is due to the base signalling rate of the Ethernet layer (100Mbps vs 1000Mbps). Lee Jone

Network backups are CPU-intensive. What are the CPUs on each end? Bob On Sat, Feb 25, 2012, at 07:54 PM, YES NOPE9 wrote: > I have run a backup over a network. It has transferred approximately 19 > GBytes in a little over 5 hours on a 100BaseT network. > [[ 19.29 GByte & ( 5:14 hours ==> 18.84K seconds ) ==> 1.204 > MByte/sec or ~ 12Mbits/sec ]] > Is this to be expected or is the network being under-utilized ( there > is no other traffic ) > gus > -

> > On Feb 25, 2012, at 10:35 PM, Bob Blick wrote: > > Network backups are CPU-intensive. What are the CPUs on each end? > > Bob > 1.83 GHz Intel Core Duos Mac OS X 2 GB RAM <== Each end > On Sat, Feb 25, 2012, at 07:54 PM, YES NOPE9 wrote: >> I have run a backup over a network. It has transferred approximately 19 >> GBytes in a little over 5 hours on a 100BaseT network. >> [[ 19.29 GByte & ( 5:14 hours ==> 18.84K seconds ) ==> 1.204 >> MByte/sec or ~ 12Mbits/sec ]] >> Is this to be expected or is the network being under-utilized ( there >> is no other traffic ) >> gus >> --

On 26/02/2012 1:54 PM, YES NOPE9 wrote: > I have run a backup over a network. It has transferred approximately 19 GBytes in a little over 5 hours on a 100BaseT network. > [[ 19.29 GByte& ( 5:14 hours ==> 18.84K seconds ) ==> 1.204 MByte/sec or ~ 12Mbits/sec ]] > Is this to be expected or is the network being under-utilized ( there is no other traffic ) > gus You should be xferring ~8mbytes/sec over 100base. sending stuff over network isnt a cpu bound task. you havent said how you are backing up i wager its your backup software doing something silly.

> > On Feb 28, 2012, at 7:19 PM, Jake Anderson wrote: > > On 26/02/2012 1:54 PM, YES NOPE9 wrote: >> I have run a backup over a network. It has transferred approximately 19 GBytes in a little over 5 hours on a 100BaseT network. >> [[ 19.29 GByte& ( 5:14 hours ==> 18.84K seconds ) ==> 1..204 MByte/sec or ~ 12Mbits/sec ]] >> Is this to be expected or is the network being under-utilized ( there is no other traffic ) >> gus > You should be xferring ~8mbytes/sec over 100base. > sending stuff over network isnt a cpu bound task. > you havent said how you are backing up > i wager its your backup software doing something silly. > Good stuff Jake I have ordered a 1000BT switch and some CAT6 cables. I want to see how that changes things. I bet you are right about backup app doing something goofy

On Tue, 28 Feb 2012 21:41:53 -0700, YES NOPE9 wrote: :: I bet you are right about backup app doing something goofy. Well with the imaging software I'm most familiar with things to look out for are - using a mapped drive letter instead of UNC or IP addressing. FTP'ing to a NAS is also feasible. There is also the possibility of poor quality cable/connectors causing resends of the data. RAM chips being flaky or if using a QNAP devices the software's drivers are not 'tuned' to that particular model. Colin -- cdb, colinKILLspambtech-online.co.uk on 29/02/2012 Web presence: http://www.btech-online.co.uk Hosted by: http://www.justhost.com.au This email is to be considered private if addressed to a named individual or Personnel Department, and public if addressed to a blog, forum or news article.

> > I have run a backup over a network. It has transferred approximately 19 GBytes in > a little over 5 hours on a 100BaseT network. > > [[ 19.29 GByte& ( 5:14 hours ==> 18.84K seconds ) ==> 1.204 MByte/sec or > ~ 12Mbits/sec ]] > > Is this to be expected or is the network being under-utilized ( there is no > other traffic ) > > gus > You should be xferring ~8mbytes/sec over 100base. > sending stuff over network isnt a cpu bound task. > you havent said how you are backing up > i wager its your backup software doing something silly. How about it is verifying each block by doing a read after write? Shouldn't need to do that on a disk to disk backup. -- Scanned by iCritical.

On 29/02/2012 3:41 PM, YES NOPE9 wrote: >> On Feb 28, 2012, at 7:19 PM, Jake Anderson wrote: >> >> On 26/02/2012 1:54 PM, YES NOPE9 wrote: >>> I have run a backup over a network. It has transferred approximately 19 GBytes in a little over 5 hours on a 100BaseT network. >>> [[ 19.29 GByte& ( 5:14 hours ==> 18.84K seconds ) ==> 1.204 MByte/sec or ~ 12Mbits/sec ]] >>> Is this to be expected or is the network being under-utilized ( there is no other traffic ) >>> gus >> You should be xferring ~8mbytes/sec over 100base. >> sending stuff over network isnt a cpu bound task. >> you havent said how you are backing up >> i wager its your backup software doing something silly. >> > Good stuff Jake > I have ordered a 1000BT switch and some CAT6 cables. I want to see how that changes things. I bet you are right about backup app doing something goofy. If your not saturating a 100mbit link, making it a 1000mbit link won't make a difference. crappy cables or switch might. without tweaking expect ~35mbyte/sec on a gig-e networ