Brian, thank you for your patience. As I am pretty new to the "backup business" it's not easy to understand how some things work.
Am 25.10.10 16:25, schrieb Brian Cuttler: > Amanda gets its parallelism by running multiple dumpers > simultaniously, multiple DLEs being dumped concurrently > to the work area, as each DLE completes it is flushed > to disk (excepting of course if you are using tape tuning > parameters but we will disregard them and stick to classic > behavor for now). As I understand this, a work area should be as large as a dump can be in a run. So the size of a specific DLE is not that important than the size of the whole dump (which depends on the levels of the dumps of the DLEs to be run on that day), means the sum of all data to be dumped to the tapes on that day, right? > If you have a 'sufficient' work area you will run multiple > dumps and you will see in your amanda reports that dump time > will be some multiple of the run or wall clock time. Currently > I imaging those values are equal 1:1. What is 'sufficient'? Is it the amount of space all DLEs take? Or just the amount of space a run of amdump takes? How can I calculate the amount of space a dump takes, as I do not know when and which DLE is full dumped and which DLEs are incrementals? Also the question of dimensioning the tapelength is not really clear for me. Should I go for tapes with the length of the largest DLE (+ some GB spare)? Or should the tapes be smaller and should I use tape_splitsize and runtapes as mentioned in my last post? At the moment my vtapes reside on a RAID5 (linux software raid). Do you think for that purpose it is sufficient to use a RAID0? The pro would be way more space and faster writes with the con of way higher risk to loose the data. But if I have real tapes there could also be a tape lost or destroyed... Another possibilty could be to do not use a RAID and to allocate the tapes across three different disks (partitions). Then I will not loose all vtapes at once. Hmmm backup is a very complicate part of IT business ;-) Thank you for your help! /Thomas
