Abstract:
Improved techniques for disaster recovery of storage service accessible through storage area networks are disclosed. Embodiments include establishing peer relationships between nodes of primary and secondary computing clusters. Configuration information is replicated in near synchronous manner from a primary cluster and used to configure a secondary cluster in a similar manner. Host data is replicated in synchronous manner for all the storage devices and LUNs within them. The secondary cluster is kept updated continuously with configuration information and user data in LUNs on storage devices such that a host may be served data from the secondary cluster when the primary cluster experiences a disaster or failure. Other embodiments are described and claimed.