Edge Computing Disaster Recovery: One of the biggest challenges facing IOT is securing IOT devices. There have been multiple incidents of security attacks targeting IOT devices in recent years. We have seen headlines of how hackers have targeted vulnerable “Things” to perpetuate attack against the IOT network or gain access to the data from the IOT devices. Whether the data from these attacks are misused or lost, it’s a financial and reputation loss for the organization. There are several types of research and new projects devising new solutions to predict and prevent such attacks from happening. In addition to data loss from a security attack, there could be situations where data failure can happen, like a natural disaster or power outages. While the discussions on security threats continue, we should also have a conversation on the disaster recovery.
IOT has been expanding into different sectors like manufacturing health care and oil and natural resources. Loss of data in these mission-critical sectors could be detrimental to the entire organization. Hence implementing a resilient IOT network with well-planned recovery is vital for the success of the IOT. Disaster recovery is a method to recover the lost data and functionality if the event of a harmful attack quickly and effectively. Traditionally the disaster recovery plan for any organization is part of the IT. When IOT architecture stores the data in the cloud of the organization, the disaster recovery is covered as part of the organization IT infrastructure.
Edge networking has been growing in popularity and is considered as the future of IOT. It is predicted that the global edge IOT market might reach up to 3.25 billion by 2025. With edge computing, the IOT data computation and analysis is moved closer to the source of data generation, near where the sensors are located. The most significant benefit of this is real-time processing and action. These data are usually found on the manufacturing floor or in the hospital floors or the oil refineries. Most cases, this will be outside the organization’s IT network. Nevertheless, these edge IOT data and assets need to be provisioned, protected, and monitored similar to the data inside an organization IT. So there should be a DR plan for these mission-critical data.
We should bear in mind that an edge IOT closer to the data collection point also means there are some constraints on the resource. There won’t be unlimited storage availability or high-power system. Also, in an organization, there could be multiple edges, and it could all be geographically distributed. This does not mean the security, data protection, asset protection, and monitoring can be less efficient for edge computing. These are mission-critical data, and a robust disaster recovery plan should be in place.
There are several ways of devising DR strategy for edge IOT. In this blog, we look at some of the prominent ones. There are some companies which provide comprehensive Edge IOT DR solution. So that can be adopted too. However, in case an organization wants to manage the DR for edge networks within their IT capacity, then this section could provide some pointers.
Using the cloud for Edge Computing Disaster Recovery
The data collected in the edge is not expected to remain in the edge; neither is it going to be flushed from the devices. Although edge networking requires data computation and analysis closer to the sensors, it has limited resource and needs to plan to send the data to the cloud. So, cloud as a disaster recovery plan for an edge should be designed and should include data backup and data protection for those from the edge networks. Also, edge data can be distributed and can come from different geolocation. Therefore, while choosing a cloud provider, support for multiple geolocations from the vendor should be evaluated. Also, there should be a plan for failovers. When the edge is sending data to the cloud, and the data center is down for any reason, there should be a failover plan.
Edge to Edge Computing Disaster Recovery
The whole purpose of the edge computing is to prevent the bottlenecks for sending the data to the cloud and have the data analysis situated closer to get the Realtime action on data. Data is transported to the cloud in batches at a scheduled time after the processing is completed usually at night. So what if the attack or disaster happens before its send to the cloud. A secondary recovery plan to back up the data to another edge in the proximity would be beneficial.
Edge – Edge, cloud – Edge Recovery
The disaster recovery plan in an edge network should also plan to build redundancies to ensure high availability. There should be plans to backup data to multiple edge device and to the various cloud in case the primary backup fails.
Another crucial consideration when planning for edge disaster recovery is communication. There are many entities at play here; there are the organizations IT, the third-party cloud provider, the edge networks. So, a successful handshake between all the parties is key to ensuring the disaster recovery is successful.
Edge Computing Disaster Recovery: The addition of the edge to the existing IOT network has complicated the DR plan. The organizations should reevaluate and should have a strategy inclusive of the edge. The conversation between the cloud vendors, edge, and organizations IT should be encouraged. And finally, with proper support from the organization management team, a robust disaster recovery plan can be in place.
- Mobodexter, Inc based in Redmond, WA builds IoT Edge solutions for enterprises applications on Kubernetes & Dockers.
- Register now at developers.paasmer.co and try a free Edge trial license of Edge Intelligence software that runs on Raspberry Pi.
- Follow our all our weekly IoT Blogs: http://126.96.36.199:8004
- Our IoT Newsletter reaches 3000+ subscribers– Subscribe Now.
- Become our affiliate partner today.