Hi!
There are a lot of finished projects which must be kept for future.
each project is a folder , which contains different kind of data in different subfolder levels
the average size of a project 2-3 Tb
There are a lot of different files inside - tiny ones, big ones , and even huge ones
the task is - sending all of projects to the cloud storage AWS S3
The question is - which way is better and convenient ??????????
I've got a couple of ideas so far:
1) pack each project in multipart tar archive by 150Gb each part. so project "xxx" looks like "xxx.tar00 xxx.tar01 xxx.tar02" etc
2) all the same but gzip packing is added (the key z in tar command). the tarball appears a bit smaller, but pack/unpack speed is a bit slower
May be somebody has faced such trivial task and could give an efficient and convenient solution. May be there are better option than tar...
and another question in addition:
there are different ways to make multipart tar archive
packing tar into file and send it via pipeline to split command like this - tar -xvf - xxx | split -d -b150G xxx.tar
or using built-in tar options like --tape-length=150G --file=xxx{00..50}.tar
but I heard that in this case there is probability NOT to unpack this archive if you have another tar version on another computer. Because different versions may have different parts delimiter format or something like this )))
Would like to hear something from professionals....
Thanx..
There are a lot of finished projects which must be kept for future.
each project is a folder , which contains different kind of data in different subfolder levels
the average size of a project 2-3 Tb
There are a lot of different files inside - tiny ones, big ones , and even huge ones
the task is - sending all of projects to the cloud storage AWS S3
The question is - which way is better and convenient ??????????
I've got a couple of ideas so far:
1) pack each project in multipart tar archive by 150Gb each part. so project "xxx" looks like "xxx.tar00 xxx.tar01 xxx.tar02" etc
2) all the same but gzip packing is added (the key z in tar command). the tarball appears a bit smaller, but pack/unpack speed is a bit slower
May be somebody has faced such trivial task and could give an efficient and convenient solution. May be there are better option than tar...
and another question in addition:
there are different ways to make multipart tar archive
packing tar into file and send it via pipeline to split command like this - tar -xvf - xxx | split -d -b150G xxx.tar
or using built-in tar options like --tape-length=150G --file=xxx{00..50}.tar
but I heard that in this case there is probability NOT to unpack this archive if you have another tar version on another computer. Because different versions may have different parts delimiter format or something like this )))
Would like to hear something from professionals....
Thanx..