Providing rapid and safe recovery within desired service levels is a key part of building mission-critical Exchange deployments. Microsoft has continued support for previous methods and strategies for performing backup and restore operations, but has evolved backup support in Exchange 2000/2003 to support new features such as multiple storage groups and databases on a single Exchange server. With proper planning and the following recommended best practices, your organization can be prepared in the event of Exchange Server data loss. This data loss can span a wide spectrum from a complete server catastrophe to simply performing an individual message recovery. In this chapter, after having laid the technology foundation in Chapter 5, I will discuss some recommended best practices for successfully implementing disaster recovery for your Exchange deployment. Of course, these best practices must be tailored to your organization s specific needs. In addition, while many of the best practices developed for previous versions of Exchange Server still apply to Exchange 2000/2003, many of them are evolving. Finally, since Exchange 2003 is a relatively new product and, thus, not widely deployed, many best practices for Exchange 2003 are yet to be developed and realized (although, they should not vary drastically from Exchange 2000). With that said, the suggestions and tips that I outline here will give Exchange system managers a jump start in developing practices and procedures that will help ensure their Exchange deployments are both highly available and recoverable.