Data Transfer
Data Mover Node
Serious moves of data to or from SciNet should be done from the datamover1 node. From any of the interactive SciNet nodes, one should be able to ssh datamover1 to log in. This is the machine that has the fastest network connection to the outside world (by a factor of 10; a 10Gb/s link as vs 1Gb/s).
Transfers must be originated from datamover1; that is, one can not copy files from the outside world directly to or from the data mover node; one has to log in to the data mover node and copy the data to or from the outside network.
ssh
All traffic to and from the data centre has to go via SSH, or secure shell. This is a protocol which sets up a secure connection between between two sites. On top of this protocol, there are many ways to copy files.
The usual ssh protocols were not designed for speed. On the datamover1 node, we have installed hpn-ssh, or High-Performance-enabled ssh. This is backwards compatable with the `usual' ssh, but is capable of significantly higher speeds. If you routinely have large data transfers to do, we recommend having your system administrator look into installing hpn-ssh on your system.
Everything we discuss below, unless otherwise stated, will work regardless of whether you have hpn-ssh installed on your remote system.
scp
scp, or secure copy, is the easiest
scp -oNoneEnabled=yes -oNoneSwitch=yes
rsync
rsync -av -e ssh 45-tangle2/*0025.bin ljdursi@nuexport00.cita.utoronto.ca:/mnt/raid-cita/ljdursi/athDrape/45-tangle2
If the files compress well, compressing first, or transmitting with compression on (scp -C, rsync -z) can significantly enhance effective data transfer rates
What transfer speeds should I expect?
| Mode | With hpn-ssh | Without |
|---|---|---|
| rsync | 60-80 MB/s | 30-40 MB/s |
| scp | 50 MB/s | 25 MB/s |
Why are my transfer so much slower?
Numbers << less than that: could be a number of things.
- network connection between scinet and your machine - how busy is the server - how busy is the disk