A VMware environment built on two Dell servers and shared storage supports a retail business's core systems. We explain how the design helps handle failures, separate services and provide backups.
Contents
The task: reducing dependence on a single server
The retail business used several important services: Windows systems, shared files, printing, an SQL database and an online shop. It needed infrastructure that would allow these services to be restored even if a physical server failed.
BitNext built a VMware environment consisting of two Dell servers and shared storage. The services were placed on separate virtual machines, allowing them to be managed and maintained independently.
How does the solution work?
The two physical servers form a cooperating group known as a cluster. Both can access the virtual machines on the shared storage.
- Two servers provide the computing capacity. Workloads can be distributed between them, and another machine is available to restart the virtual machines from a failed server.
- Shared storage holds the virtual disks. Both servers connect to the same storage environment.
- The system is managed centrally. VMware vCenter provides an overview of the servers and virtual machines.
If hardware fails, VMware HA can automatically restart the affected virtual machines on the other server. This reduces manual recovery work; because restarting takes time, it does not mean uninterrupted operation for every service.
A separate system for each important task
| Service | Role in the business |
|---|---|
| Windows servers | Managing users, permissions and essential network services. |
| File server | Storing shared documents and controlling storage usage. |
| Print server | Managing company printers centrally. |
| SQL database | Serving business application data, with database-level redundancy. |
| Online shop | A separate Linux environment for online sales. |
Separating the services makes operations easier to understand: updates or troubleshooting for an individual system can be carried out independently.
Backup and access protection
High availability and backup serve different purposes. Failover between servers helps address hardware failures; separately stored backups may be needed after data corruption or accidental deletion, for example.
The design includes backups to an external backup server and network separation. Administration, customer traffic and database connections operate in separate network segments, controlled by firewall rules.
Snapshots help record a previous state, but they do not replace separately stored backups on their own.
What did the business gain?
- Less dependence on a single physical machine: if a server fails, another is available to restart the affected systems.
- Clearer operations: services run on separate virtual machines and can be managed centrally.
- More predictable expansion and maintenance: capacity changes and updates to individual services can be coordinated.
- Multiple layers of protection: server redundancy is complemented by backups and controlled network access.
Technical details
| Component | Configuration |
|---|---|
| Physical servers | 2 × Dell PowerEdge R750; each server has 2 × Intel Xeon processors and 128 GB of memory. |
| Shared storage | Dell EMC Unity XT 380, 10 TB SSD, RAID10; iSCSI connections to both servers. |
| Virtualisation | VMware vSphere 8.0, ESXi and vCenter Server; HA and DRS features. |
| Windows environment | Active Directory, DNS and DHCP; file services with DFS replication and quotas, central printing with Group Policy. |
| Database and online shop | SQL Server 2022 AlwaysOn; a separate Linux LAMP environment, reverse proxy and CDN. |
| Backup and networking | Acronis backups to an external backup server; separate networks for administration, customer traffic and the database. |
Want a more reliable server environment for your business? Tell us which systems you use, and we will help plan the right infrastructure.