ensignandrew
Technical User
Hello all,
I'm running into an issue where I'm getting slow throughput when backing up our backup server to tape.
Here's some background info:
- We're trying to backup about 7.3 TB
- Average throughput is about 450 MB/min
- Server is running Windows Server 2003 SP2
- Backup software is ARCserve 12 SP1
- Server is connected to the RAID array with a 3 Gb/s link
- Server is connected to the tape library with a 4 Gb/s fibre link
- Tape drives are LTO 4 and can run about 114 MB/s
I have already updated the drivers & firmware of the RAID and fibre adapters, the tape library, and the tape drives. I also recreated the backup job in case the original was corrupt. The job seems to start with decent speed (around 1,500 to 2,000 MB/min) and will be up around 3 TB complete the following morning, but when I check again the next day, the throughput will be down around 400 MB/min and the amount of files completed will only be about 100 GB. It's almost like it's getting halfway and then some part of the job is restarting.
I have a second job running that backs up our other physical servers. That one completes with an average throughput of between 1,200 MB/min to 1,500 MB/min even though it's going over the network.
I have a case open with CA, but I was hoping someone here might be able to throw out suggestions as to what could be causing the slowdown or what I might try to speed things up.
Thanks
Andrew
I'm running into an issue where I'm getting slow throughput when backing up our backup server to tape.
Here's some background info:
- We're trying to backup about 7.3 TB
- Average throughput is about 450 MB/min
- Server is running Windows Server 2003 SP2
- Backup software is ARCserve 12 SP1
- Server is connected to the RAID array with a 3 Gb/s link
- Server is connected to the tape library with a 4 Gb/s fibre link
- Tape drives are LTO 4 and can run about 114 MB/s
I have already updated the drivers & firmware of the RAID and fibre adapters, the tape library, and the tape drives. I also recreated the backup job in case the original was corrupt. The job seems to start with decent speed (around 1,500 to 2,000 MB/min) and will be up around 3 TB complete the following morning, but when I check again the next day, the throughput will be down around 400 MB/min and the amount of files completed will only be about 100 GB. It's almost like it's getting halfway and then some part of the job is restarting.
I have a second job running that backs up our other physical servers. That one completes with an average throughput of between 1,200 MB/min to 1,500 MB/min even though it's going over the network.
I have a case open with CA, but I was hoping someone here might be able to throw out suggestions as to what could be causing the slowdown or what I might try to speed things up.
Thanks
Andrew