disaster recovery as a cloud service

Disaster Recovery as a Cloud Service

Chao LiuSUNY Buffalo

Computer Science

A typical DR service works by replicating application state between two data centers; if the primary data center becomes unavailable, then the backup site can take over and will activate a new copy of the application using the most recently replicated data.

How is DR Done Today?

Recovery Point Objective The RPO of a DR system represents the point

in time of the most recent backup prior to any failure.

Recovery Time Objective The RTO is an business decision that specifies

a bound on how long it can take for an application to come back online after a failure occurs.

Disaster Recovery Requirements

Performance

Consistency

Geographic Separation

Disaster Recovery Requirements

State replication can be done at one of these layers: Within an application Per disk or within a file system For the full system context

Replication at the application layer can be the most optimized, only transferring the crucial state of a specific application.

Disaster Mechanism

The level of data protection and speed of recovery depends on the type of backup mechanism used and the nature of resources available at the backup site.

1. Hot Backup Site 2. Warm Backup Site 3. Cold Backup Site

Disaster Recovery

Provides a set of mirrored stand-by servers These servers are always available to run Provide minimal RTO and RPO Synchronous replication This form of backup is the most expensive Largest impact on normal application

performance

Hot Backup Site

Depend on the necessary RPO

Synchronous or asynchronous

It may take minutes to bring them online

Slows recovery, but also reduces cost

Warm Backup Site

Data is only replicated on a periodic basis

Servers are not readily available

A delay of hours or days, resulting in a High RTO

Cold Backup Site

Detect when a disaster has occurred is a challenging problem.

Cloud based systems can simplify this problem by monitoring the primary data center from cloud nodes distributed across different geographic regions.

Failover and Failback

During normal operation, the system stays in Replication Mode.

When a disaster occurs, the system enters Failover Mode.

DR as a Cloud Service

Cost Breakdown

99% Uptime Cost

Cost of Adding DR

Case Study: Multi-tier Web App

Cost Breakdown

99% Uptime Cost

Periodic Backups

Case Study: Data Warehouse

Cloud computing can facilitate disaster recovery by significantly lowering costs

The benefits of virtualization, while not necessarily specific to cloud platforms, still provide important features for disaster recovery

Benefits of the Cloud

Handling Correlated Failures

Revenue Maximization

Challenges for the Cloud Provider

Network Reconfiguration

Security & Isolation

Mechanism for Cloud DR

Cloud computing platforms are an excellent match for providing disaster recovery services due to their pay-as-you-go pricing model and ability to rapidly bring resources online after a disaster.

Better understand what features and optimizations must be included within the cloud platform itself, and to explore the tradeoffs between cost, RPO, and RTO in a cloud DR service.

Ongoing Work and Conclusion

Thank You