What You Need to Know About Data Replication
When you are creating a highly available infrastructure for your virtual machines, replication is one of the most powerful tools that you can use to protect your virtual infrastructure. Replication enables you to keep one or more copies of your virtual machines (VMs) on secondary servers. Those servers can either be on-premises, in a geographically separate location, or even in the cloud. Replication provides very good recovery time objective (RTO) and recovery point objective (RPO) values because it enables you to have one or more copies of your VMs in a ready-to-start state. Fast RTO times are one of the main reasons many businesses use replication to protect their tier 1 mission-critical applications.
Replicas are typically created by using a snapshot or backup of an existing VM. The replication initialization process makes a copy of the VM running on the source host and then uses that copy to create a full replica of the VM on the target host. The replica is usually stored in its native virtualization format. Then, as the VM continues to be used on the source host, the changed data blocks in the source VM are replicated to the target VM replica.
Although the basics of replication are fairly straightforward, there are several factors that can influence the efficiency and effectiveness of the replication process.
Replication is typically set up to take place at regular intervals. The frequency at which the replication process runs determines the RPO (i.e., the point-in-time at which you can restore your VM) that you can achieve. If you replicate changes only once per hour, the best RPO is one hour. Likewise, shorter intervals, such as every five minutes, give you more granular RPOs. The interval that you use depends on your organization’s recovery needs, as well as the activity level and workload of each VM. Some replication technologies provide change block tracking in the source VM with extremely frequent replication intervals to enable near-continuous data protection.
Data Compression and WAN Acceleration
In many cases, replication needs to happen over slow WAN or Internet connections, which can increase the time required to apply the source replication changes to the replica target. In these situations, making use of data compression can be one of the most effective ways to reduce the data volume that’s being transferred. In addition, WAN acceleration technologies can filter out the transmission of unnecessary data, such as duplicate data blocks, and swap file blocks to further compress replica traffic.
Security is a major concern for most organizations, and if your replicas are located across Internet connections, then in-flight data protection can be a vital component to protecting your VMs. Both the replica backup used to initialize replications and the replica traffic from the host to the target should be encrypted to prevent any possible unauthorized access.
Today, many businesses are using hybrid cloud implementations to extend their on-premises infrastructure into the cloud. In hybrid cloud scenarios, the cloud can serve as your replication target, providing high availability and disaster recovery for your VMs. To enable hybrid cloud replication, the cloud target needs to be able to support the VM replication technology that your organization is using.