I have a question about job structure. First off I have five backup servers(dell 2650 and 2850's)and the following tape libraries
All run w2k3 sp1 and arcserve 11.5 SP2
Tape Servers
PE 2850 and pv 136t -3 LTO2's
PE 2850 and pv 136t -3 LTO2's
PE 2650 and pv 136t -2 LTO's
PE 2850 and pv 132t -2 LTO's
Disk staging server
PE 2850 with pv 220 4TB of disk storage
Most of the libraries are broken up in 2 or 3 jobs each with approx 10-25 servers in it. The jobs are GFS in a 5 day rotation with the media in a 5 week offsite rotation. Each job has it's own media pool. Jobs have exploded in size in the last six months. So the once adeqaute media pool of 15 slots is not any longer. Media is running short. The disk staging server is not utilizing its space and I feel is setup as a single point of failure. I am looking to restructure the jobs. Possibly one solution I am looking at is create one media pool for an entire library. Remove all old jobs and pools. Add each server as its own job or create one job out of the three old jobs. I would use media maximization with GFS rotation. My reason for this is to simplify media managment and maximize media storage. In addition I would have more flexibility with scheduling servers for backups if I used the 1job=1 server approach. I would also take disk from the staging server and put it into the tape servers, giving each server 1TB for local disk to tape backups. It doesn't make much sense to have all the disk storage in one location and not connected to any tape system. If anyone has any insight to these issues it would be greatly apperciated. The overhead to manage this backup system is getting annoying. Thanks.
All run w2k3 sp1 and arcserve 11.5 SP2
Tape Servers
PE 2850 and pv 136t -3 LTO2's
PE 2850 and pv 136t -3 LTO2's
PE 2650 and pv 136t -2 LTO's
PE 2850 and pv 132t -2 LTO's
Disk staging server
PE 2850 with pv 220 4TB of disk storage
Most of the libraries are broken up in 2 or 3 jobs each with approx 10-25 servers in it. The jobs are GFS in a 5 day rotation with the media in a 5 week offsite rotation. Each job has it's own media pool. Jobs have exploded in size in the last six months. So the once adeqaute media pool of 15 slots is not any longer. Media is running short. The disk staging server is not utilizing its space and I feel is setup as a single point of failure. I am looking to restructure the jobs. Possibly one solution I am looking at is create one media pool for an entire library. Remove all old jobs and pools. Add each server as its own job or create one job out of the three old jobs. I would use media maximization with GFS rotation. My reason for this is to simplify media managment and maximize media storage. In addition I would have more flexibility with scheduling servers for backups if I used the 1job=1 server approach. I would also take disk from the staging server and put it into the tape servers, giving each server 1TB for local disk to tape backups. It doesn't make much sense to have all the disk storage in one location and not connected to any tape system. If anyone has any insight to these issues it would be greatly apperciated. The overhead to manage this backup system is getting annoying. Thanks.
