*6.11.3 Failover and failback procedures*

In the event of a disaster, the failover procedure excludes failed resources and redirects workloads to a secondary site. Client-transparent procedures and fast IP failover requirements are two main challenges raised by this context. On the other hand, the cloud environment recovers from the disaster, application control has to be reverted to the original site. For this purpose, bidirectional state replication must be supported by the DRaaS mechanism. A portion of data may be lost because of the disaster in the primary site and new data will be created in the backup site. Therefore, one major challenge is how to determine new and old data which must be resynchronized to the primary site [24].

## *6.11.4 Disaster monitoring*

In the case of a disaster, the sooner the failure is detected in either the primary site or the backup site, the better RTO. So, the challenge is how should the status of cloud be monitored and how a disaster can be detected as soon as possible [27].

### *6.11.5 Resource scheduling*

The number of cloud-based services is increasing day by day and so has increased the complexity of cloud infrastructures. Hence, resource scheduling is a critical issue in modern cloud environments. This issue is more crucial for cloudbase platforms since they face unpredictable incoming traffic and have to consider a *Network Function Virtualization over Cloud-Cloud Computing as Business Continuity Solution DOI: http://dx.doi.org/10.5772/intechopen.97369*

variety of catastrophic situations. Building on this, more efficient resource scheduling techniques are needed in order for current DRaaS platforms to be optimized.
