We're new to NAS and SAN but need to set up a centralized backup possibly, using Veritas Backup Exec.
What is the difference and best option in setting them up to back up about 10 servers in one location and three in another? Also, we're implementing Fibre for all locations.
Click here for part 1.
Common backup methods:
- Integrated tape drive in each server: This is a design which that uses individual tape drives in each server, which have enough capacity to store all the data that the server has access to.
- Backup over the corporate network to tape drive connected to a dedicated backup server: This policy allows the backup server to back up the data on the application server's behalf. By using this policy you can use a single high-capacity tape device rather than including a tape drive in every server. This is a great method but it requires someone to change the tapes when one becomes full.
- Backup over the corporate network to robotic tape library connected to dedicated backup server: This is the most common backup policy used today for large corporations. It allows data to travel across the network to the backup server that is connected to a large tape library. The library includes robotics to change out tapes as when they become full.
- LAN-less backup to shared tape library over SAN: Using the SAN as the data path for backup eliminates the need for the backup stream to use the corporate LAN. Using the SAN also removes the time constraints backup imposes on production applications. You can back things up any time day or night without worrying about impacting affecting the performance of the production LAN.
- Server-less backup to shared tape library through SAN: Server-less backup is new. You need a SAN in order to accomplish server-less backup. This method uses a new SAN backup protocol called e-copy to move data directly from disk drives to tape libraries that removes the servers from the data path.
- Disk-to-disk backup: Since disks are becoming less expensive every day, some companies are foregoing tape drives all together. They back up the data from the primary disk subsystem to a less expensive array of disks. This method enables very fast restore times.
- Image copy in SAN: SAN-based image copies is a method of using intelligent storage arrays to create exact duplicates of your data on another disk inside the array. The image copy can then be used to recover your data instantly in case of primary disk failure or data corruption. Image copies can also be used as backup source volumes. Instead of using the original production disk, you shut down the application only long enough to create the image. The application is then brought back up, and the image is used as the source for backing up to tape through the SAN. This almost eliminates the backup window, which you can read more about in the earlier section "The backup window."
- SAN data replication/remote backup: Another backup method that is commonly used as part of a disaster recovery solution is to move data from one site to another through an extended SAN. This keeps data safe at a remote contingency site in case of fire or other disasters.
Editor's note: Do you agree with this expert's response? If you have more to share, post it in one of our .bphAaR2qhqA^0@/searchstorage>discussion forums.
This was first published in April 2003