A Case Study from a Home Improvement
and Repair Retailer
How to Maintain High Performance During Marketing Campaigns and Optimize IT Infrastructure Scaling
About the Client
A growing chain of home improvement and repair stores. At the start of the project, the company operated 45 retail locations and was actively expanding its e-commerce division. Its core business strategy was to be "your one-stop shop for all things home improvement and repair," which necessitated a vast product assortment and complex logistics. Management viewed the IT infrastructure as a strategic asset, critical for supporting the seamless operation of its physical stores, warehouses, and online storefront.
Initial State Before the Project
The company's infrastructure was built on a Proxmox VE virtualization platform, deployed several years earlier on HP server hardware. The total estate comprised 28 physical servers, split between a primary data center and a disaster recovery site.
Business-critical systems ran in virtualized environments, including:
- The product catalog and pricing database management system (DBMS)
- The ERP system
- The warehouse management system (WMS)
- A suite of customer-facing web services
The virtual machine configuration was far from optimal. Most VMs were over-provisioned with resources "just in case," while mission-critical systems simultaneously suffered from performance bottlenecks. The backup solution, built on Proxmox's native tools, failed to meet the required recovery time objectives (RTO).
Challenges
The existing IT infrastructure could no longer keep pace with business growth, creating a set of critical issues that began to directly impact daily operations and the company's bottom line.
The core database management system (DBMS) for the product catalog and pricing became unstable, particularly during peak loads. Delays in updating prices and inventory levels reached 15-20 minutes, leading to point-of-sale conflicts, customer dissatisfaction, and direct financial losses. Under maximum load, the average response time of the ERP system was 3-4 times higher than acceptable service-level agreements.
Adding new servers to the cluster required complex reconfiguration and did not guarantee stable operation. The company found itself unable to scale resources quickly for seasonal promotions (such as spring sales) or the opening of new stores. The ability to launch new digital initiatives was stifled by these technological constraints.
Adding new servers to the cluster required complex reconfiguration and did not guarantee stable operation. The company found itself unable to scale resources quickly for seasonal promotions (such as spring sales) or the opening of new stores. The ability to launch new digital initiatives was stifled by these technological constraints.
The legacy backup system could not meet the required recovery time objectives (RTO). A full restoration could take up to 12 hours, which was unacceptable for a 24/7 retail operation.
Despite an average physical server utilization rate of just 35-40%, key business applications were starved for computational power due to the suboptimal allocation of virtual resources. The company was spending an estimated X annually on maintaining this over-provisioned infrastructure without any tangible performance gains.
Updating the virtualization platform required a full 6-8 hour service outage, presenting an unacceptably high business risk. System administrators were spending up to 30% of their time on routine monitoring and maintenance tasks just to keep the virtual environment running.
Project Goals and Objectives
The identified issues demanded a comprehensive solution—not merely a virtualization platform replacement, but the creation of a new technological foundation for future business growth.
The project's key goals were:
- To reduce operational expenditures for managing and supporting the virtual environment.
- To ensure service reliability and minimize recovery times in the event of failures.
- To establish a unified management framework for the entire IT infrastructure—from physical hardware to virtual environments.
- To enhance the performance and stability of mission-critical business systems, specifically the product catalog DBMS and the ERP system.
- To build a scalable and agile IT infrastructure capable of rapidly adapting to the needs of a dynamic retailer, including the launch of new digital initiatives.
To achieve these goals, the client needed to accomplish several key objectives:
- Migrate from Proxmox VE to a business-ready platform without disrupting ongoing business operations.
- Implement a solution for the centralized management of the entire physical server fleet.
- Reduce the time IT staff spent on routine virtualization management tasks.
- Deploy an efficient backup and disaster recovery solution with a target recovery time objective (RTO) of no more than X hours.
- Optimize resource allocation across virtual machines with guaranteed performance for priority systems.
- Create an infrastructure that allows computational resources to be scaled within a 24-hour window to support seasonal peaks and new business initiatives.
Solution Requirements
To address its infrastructure transformation needs, the company deployed a combined solution using ISPsystem's VMmanager for virtualization and DCImanager for infrastructure control.
The key deciding factors for this choice were:
A Unified Ecosystem — the seamless, out-of-the-box integration between VMmanager and DCImanager, access to centralized technical support, and the ability to establish end-to-end management workflows.
Legacy Hardware Support — full compatibility with the existing server fleet, no vendor lock-in for future hardware procurement, and the ability to operate in a heterogeneous environment.
Ease of Deployment and Use — an intuitive user interface, rapid deployment capabilities, minimal staff training requirements, and comprehensive documentation.
Cost-Effectiveness — a transparent licensing model with no hidden fees, providing predictable total cost of ownership over the long term.
Business-Grade Functionality — built-in tools for operational task automation and the ability to scale in line with future business demands.
Following a thorough evaluation, the combined VMmanager and DCImanager platform demonstrated the strongest alignment with both the technical specifications and the strategic business needs of the company.
Project Implementation
The implementation of VMmanager and DCImanager was executed in several carefully planned phases to minimize business risk.
During the preparatory phase, the client's IT team, with support from ISPsystem's engineers, conducted a thorough assessment of the existing infrastructure and developed a detailed, phased migration plan. To mitigate risk, test environments were deployed to rehearse all migration procedures, and key IT department personnel received comprehensive training.
The deployment of the DCImanager platform provided the client's team with the ability to rapidly onboard and inventory all physical servers, as well as configure comprehensive hardware monitoring.
The next step was the virtualization infrastructure migration.
Following the successful deployment of VMmanager, the IT department designed optimized virtual machine configurations for the critical systems: the DBMS, the ERP system, and the customer-facing web services. The phased migration of virtual machines was accompanied by rigorous testing after each step, culminating in the configuration of high-availability mechanisms.
In the final phase, the team automated backup and recovery procedures and established a detailed, role-based access model for administrators, ensuring the secure and efficient management of the new infrastructure.
The project was completed on schedule, a direct result of meticulous preparation and the seamless collaboration between the client's internal specialists and the vendor's support engineers.
Key Features of the Implemented Solutions
DCImanager was integrated to automate the management of the physical infrastructure, providing the following key capabilities:
Unified Remote Server Management
The platform offers extensive remote server management features, simplifying administration and increasing its efficiency. It provides effective monitoring of the physical health of servers, enabling the proactive prevention of incidents. This automation reduced the time required for routine administrative tasks from several hours to just minutes.
Hardware Health Check
The platform ensures control over the state of the entire physical infrastructure—server and network equipment, UPSs, and PDUs—across all connected locations. Real-time monitoring with instant notifications helps resolve issues promptly and avoid unplanned downtime.
Network Equipment and Virtual Network Management
DCImanager enables control of any switch and the inventory of network resources. Adding a switch, configuring ports, and setting up VLANs can be done in a few clicks. The documentation of network settings, IP address accounting, and assignment are all automated. This gave the client the ability to quickly connect new retail locations to the corporate network without requiring on-site visits.
Hardware and Component Inventory
The asset management module allows for the tracking of servers, switches, routers, UPSs, and other devices. Custom equipment types, such as patch panels and cables, can be added. The platform automatically detects components installed in servers (CPUs, RAM, hard drives), eliminating the manual work of populating asset records with detailed specifications. The company gained an accurate database of all equipment, simplifying procurement and maintenance planning.
High Scalability
VMmanager's architecture is built for massive scalability. A single installation can manage 56,000+ VMs, 50+ clusters, and 350+ physical servers. It can be scaled further by adding more platform installations. This capability laid the foundation for multi-year IT infrastructure growth without the need to change the technology stack, giving the client confidence that their IT would match business growth for the next 5-7 years.
Ensuring Service Continuity
Features like automated load distribution (DRS) and the ability to modify VM parameters without downtime allowed the business to perform planned maintenance and scale computing power for resource-intensive tasks without stopping services or applications. For example, the team could increase resources for the online ordering system during peak hours without interrupting customer service./p>
Centralized Infrastructure Monitoring
The built-in monitoring system provided the IT department with a single pane of glass for tracking key metrics of the virtual environment. Time spent on diagnostics was significantly reduced, and proactive alerts on threshold violations allowed the team to prevent failures before they occurred.
Platform and Cluster-Level High Availability
Thanks to its "Unbreakable clusters" feature, VMmanager guarantees a high level of fault tolerance. If a node fails, the platform automatically migrates VMs to healthy servers within seconds, meeting the requirements for uninterrupted operational processes. In practice, this allows the company to ensure the continuous operation of its most critical applications.
On-Demand Scaling for Peak Loads
The mechanism for seamlessly adding new hosts to a cluster with automatic load balancing provided the client with a tool for rapid response to traffic spikes. The IT department can now scale capacity up quickly and without downtime, while the platform's built-in load balancer ensures even resource distribution. During marketing campaigns, the IT team can promptly add new servers to the cluster, enabling the infrastructure to handle a multiple-fold increase in load on the e-commerce store.
Results and Future Plans
The implementation of VMmanager and DCImanager enabled the company to achieve significant operational and financial results, establishing a strong technological foundation for future growth.
Key Outcomes:
Reduced Operational Expenditure
Total costs for managing and supporting the IT infrastructure decreased by 45% year-over-year.
Increased Operational Efficiency
IT staff effort spent on routine virtualization and physical hardware management tasks was reduced by 60%.
Enhanced Performance
Response times for key systems, including the DBMS and ERP, improved by 4-5x. IT service downtime was minimized—the client has reported zero critical incidents impacting business processes in the last six months.
Improved Fault Tolerance
Service recovery time after failures was slashed from 12 hours to just 5 minutes. Automated load redistribution between servers guaranteed 99% availability for mission-critical systems.
Looking to the future, the client plans to integrate another ISPsystem solution—BILLmanager. This will be used to create a self-service portal for business units, which will accelerate the deployment of test environments and reduce the time-to-market for new digital services.