Data Domain has had a good run in the backup space because their philosophy was to add technology behind existing backup infrastructure making it easy to fit into the backup space and by reducing the backup data footprint (because that is where a large percentage of duplicate data lives) they could drive the cost of disk and tape much closer, making disk a more economical backup target. However, with data increasing at such a rapid rate on primary storage, existing backup infrastructures can't keep up and it is time to do something differnt. Additionally, a number of competitive companies have talked about putting de-duplication on primary stroage to make an even more significan impact on the storage growth and costs of new storage (and in today's economy, this is a must). If vendors are putting de-duplication on primary storage, then the question is, why would I back it up to a device that uses a different de-duplication algorithm? The performance and network impact become an expense and unnecessary if I a company that has de-dupliation in primary storage and backend storage. At the end of the day, de-duplicatin is a feature that needs to be leveraged across all of your storage tiers and Data Domain doesn't offer storage at all the tiers, rendering them, over time, not useful.
