Designing and implementing a hyperconvergent cluster for the medical services startup.

Client:

Startup in Healthcare


Case review:

The Client is a young fast-growing startup, providing a free internet service, allowing users to quickly and efficiently find the necessary doctor and make an appointment online.


Task:

Design and deploy a hyper-converged failover cluster that combines storage based on locally installed hard drives; minimize software costs


Technologies:

  • Cisco ASA Firewall
  • Cisco Switches
  • HP Servers
  • Microsoft Hyper-V Server
  • StarWind Virtual SAN
  • Ubuntu Server OS


Result:

To meet the objectives, a two-node failover cluster was deployed. Cluster based on Hyper-V Server 2016, a free version of OS from Microsoft, and HP servers, allows us to fully utilize the power of Virtualization technology.

The scenario of building this failover cluster might be interesting to companies with limited budgets, but with the need for high availability of infrastructure. As a storage system, we decided to use a shareware product StarWind Virtual SAN, which allows us to organize the fault-tolerant storage based on disks locally installed in servers.

Ubuntu Server has been selected as the OS to run "guest VMs". Guest VMs running in the Hyper-V virtual environment are duplicated, and all necessary software components required for correct operation of the "service" are run in multiple instances of guest VMs.

The network infrastructure is based on Cisco equipment. All components of the network are duplicated. This cluster configuration allows the Client to survive the loss of one of the physical nodes, failure of a part of the network equipment or failure of the "Guest OS" and allows keeping the "service" operation at a high level.