High availability (HA) refers to a system’s ability to keep running and operating without disruption. In the event of an outage, there are two primary failover modes: active-active and active-standby.
Yet what if a massive outage or disaster occurs, taking down an entire site?
One of the breakthrough use cases for the Resilio Active Everywhere platform is hot site disaster recovery, where all sites (2 or more) are active. In the context of multi-site HA for Disaster Recovery, all sites in the Resilio model are active—everywhere.
Resilio synchronizes files in near real-time and bi-directionally across all sites concurrently. For use cases like hot disaster recovery, all files are always available and accessible across all sites. Depending on your implementation, using Resilio, businesses can meet sub-five-second RPOs (Recovery Point Objectives) and RTOs (Recovery Time Objectives) within minutes of an outage.
Another example is simple failover to shared storage within a site. If any system running a Resilio Active Everywhere Agent fails for any reason, Resilio enables a standby system to quickly resume operations and continue the replication job. This is an active-standby mode, which means your job continues to run and intelligently restarts after pausing for failover.
Finally, a third scenario is a cloud-hosted server instance (or another system) running a Resilio Active Everywhere Agent attached to a cloud storage service like Amazon S3. If that instance fails, Resilio enables a standby agent to resume operations quickly. Resilio enables this type of active-standby HA for any cloud object storage, including Amazon S3, Microsoft Azure Blobs, and Google Cloud object storage.
High availability is easy to configure using Groups. Before getting into the specifics of High Availability groups, let’s discuss the concept of groups in general. There are three types of groups.
Standard
Standard Groups include multiple Resilio Active Everywhere Agents. This simplifies applying policies and assigning multiple Agents to Jobs.
High Availability
Agents in this group are also assigned to Jobs but run in an active-standby failover mode. If an Agent fails, another Agent takes over.
Gateway Scale-Out
Agents in this group support advanced active-active high availability when used with File Caching Jobs–and deployed as File Caching Agents at remote sites. We also refer to these Agents as File Caching Gateways.
High Availability (HA) Groups
High Availability groups simplify file-based data management by also representing groups of Resilio Active Everywhere Agents protected by active-standby high availability. Once an agent is added to the High Availability Group, policies, job assignments, and other settings are easily applied collectively to all Agents in the group.
So why High Availability? Resilio is peer-to-peer, and P2P does a great job routing around failures to keep files flowing, synchronized, and accessible—without a single point of failure.
But what if a Resilio agent is connected to an S3 bucket—and the agent attached to the bucket fails? In a stand-alone deployment, no other agents would have access to the S3 storage until someone manually intervened and remounted the storage. With High Availability groups, you can have standby agents automatically start and resume the job in case the primary agent fails.
For example, let’s say a synchronization job was running between an on-premises network attached storage (NAS) and an Amazon S3 bucket. For whatever reason, the Resilio Active Everywhere Agent in AWS mounted to the object storage fails. Using High Availability groups, the standby Agent would resume operations and continue synchronizing the files instead of having to restart the job.
High Availability groups ensure that if an agent fails during the synchronization job, there’s an efficient, timely, and automatic recovery, ensuring the job continues without user intervention.
It makes sense to deploy a High Availability group when you need high availability to back-end storage. One thing to note: Agents in the same group should run the same operating system. A best practice is to keep Linux agents in one group, Windows agents in another group, and so forth. To learn more about High Availability groups, please see the Resilio Help Center article on High Availability Groups.
Active-Active High Availability Using Gateway Scale-out Groups
In advanced scenarios where file caching agents are deployed as caching gateways, the Gateway Scale-out Group may be used. This group includes Resilio Active Everywhere Agents assigned to file caching jobs—and used as file caching gateways. In this scenario, multiple Resilio Agents may be deployed to achieve both high availability and scale-out performance.
Each gateway is active. If one of the agents fails, another takes over. Thus, failover is active-active. Optionally, customers may use a global namespace or load balancer such as DFS and other popular tools to enable seamless failover across nodes.
Schedule a Demo
Get your organization active everywhere with High Availability Groups—see how with a personalized demo.
Complete this form and pick a slot on our calendar. Our experts are ready to learn your use case and provide a live demo of the Resilio Active Everywhere Platform.
