LanCloud Disaster Recovery as a Service

LanCloud Disaster Recovery as a Service is a cloud service of fast disaster recovery of operation of virtual servers in a reserve data processing center of LanCloud in case of a partial or total failure of the main DPC of the Customer.

2018

On September 3, 2018 the LanCloud company announced start of a cloud service of disaster recovery of servers - Disaster Recovery as a Service based on Veeam Backup & Replication and Cloud Connect.

As it was reported, most of customers, even having failover clusters of virtual machines, and regularly executing backup of all data, critical for business, do not pay sufficient attention to the plan of disaster recovery (DRP – Disaster Recovery Plan). And, having faced any serious failure, for example, failure of a cluster of virtual servers, failure of a productive storage system or other technogenic catastrophe of DPC, including withdrawal of the equipment by law enforcement agencies, the company appears in emergency situation to which it was not in advance ready.

Usually at such moment it becomes clear that even having all necessary backup copies if they were stored on other equipment in other data center, at the company:

• there is no spare equipment on which it would be possible to start all server framework;
• even if spare equipment is present, recovery time will be several days or weeks, depending on amount of data;
• the relevance of data in backup copies is several days or weeks until failure, and even after recovery completion of the procedure, all information which collected from the moment of the last backup will be lost.

Such events often lead to multimillion losses for business and, as a result, to bankruptcy of the company.

To be ready to incidents of this sort, each company should have in advance prepared DRP plan. The DRP plan or the plan of disaster recovery is the document describing the sequence of actions which should be executed to recover work of IT infrastructure in case of this or that failure or accident in the main DPC. Two major parameters of the DRP plan are parameters: RTO and RPO.

RTO (Recovery Time Objective) is time necessary on complete recovery of work after failure.

RPO (Recovery Point Objective) is a recovery point, or in a different way - on what timepoint before failure all information of the company will be recovered.

In turn, each business should define what RTO and RPO parameters are for it acceptable at approach of unfavorable events. The less these indicators - the better. But if it is about providing RTO and RPO no more than several minutes, then it is obvious that simple backup will not help here.

For the purpose of providing to Customers the ready tool allowing to achieve operational disaster recovery of IT infrastructure with minimum loss of data, LanCloud brought Cloud DRaaS service (Disaster Recovery as a Service) to the market.

According to the company, Cloud DRaaS is a cloud service of fast disaster recovery of operation of virtual servers in a reserve data processing center of LanCloud in case of a partial or total failure of the main DPC of the Customer. The Cloud DRaaS service provides permanent asynchronous replication of all virtual servers from the main DPC of the client in reserve DPC of LanCloud. In case of failure of the main DPC, the Failover procedure is activated, and all virtual machines automatically begin to work in LanCloud cloud. The Cloud DRaaS service provides indicators of RTO and RPO no more than 15 minutes.

Cloud DRaaS is constructed based on the solutions Veeam Backup & Replication and Veeam Cloud Connect. In DPC of the Customer the Veeam BackUp & Replication server which replicates virtual machines in Veeam Cloud Connect in LanCloud cloud is installed. In case of failure, the Customer activates the procedure of abnormal switching, and all virtual servers are started in LanCloud cloud. At the same time, L2 a VPN tunnel to the customer's infrastructure automatically is established that allows all servers to remain available on internal network to the same IP addresses.

Abnormal start of virtual servers in a cloud in case of refusal local DPC

At recovery of the main DPC, the administrator can start the procedure of the return switching – "Failback", and all servers will be back otreplitsirovana and are let into DPC of the Customer with up-to-date data. In addition to the procedure of abnormal switching, to customers functions of test start of virtual servers in a cloud without influence on productive infrastructure and also function of recovery of separate files and applications from cloud remarks are available, to similarly normal backup.