Moving from in-house IT to third-party hosting services: Case study
A supplier of civil engineering and building materials moved its IT infrastructure to a third-party data centre to resolve its in-house IT problems, including cooling and flooding issues.
For 10 years, a civil engineering and building materials supplier with an in-house IT model faced power and air-conditioning failures as well as flooding due to water leaks.
But as the business grew and data centre operational problems worsened, the company decided early this year to use third-party hosting and colocation services and move away from its traditional in-house IT infrastructure strategy.
The civil engineering and buildings materials supplier WTB Group opted to relocate its computing infrastructure to a secure data centre environment with redundant power and cooling and protection from environmental risks.
“The main reason was uptime; we needed to improve availability because all our business operations now rely on IT services,” said Tim Brice, WTB’s IT manager. “With offices and stock locations in the UK, Ireland, Australia and Dubai, and with expansion plans, a 24/7 IT operation became a necessity.”
Brice was also conscious of the budget restrictions. “We didn’t want a fully managed hosting service because contracts can be expensive and not cost-effective for the size of our IT operation,” he said.
As the company is in the process of achieving ISO27001 compliance in order to meet tougher tender requirements from public and private sectors, Brice also wanted a third-party data centre that was ISO27001-compliant.
After considering a number of hosting and colocation service providers, including British Telecom, WTB Group selected Databarracks.
An existing relationship between WTB and Databarracks made the choice of vendor much easier. “We were already using Databarracks for backup and disaster recovery,” Brice said. “Having our IT infrastructure and backup infrastructure at the same location will help in quicker and more efficient backups, server and data replication and recovery in the event of a disaster.”
Problems of in-house IT
Brice explained why disaster recovery is so important for his company. In 2008, its storage area network (SAN) with four terabytes of data and 36 virtual servers experienced problems.
A drive failed and when it attempted to recover itself, the SAN suffered corruption across the volume group, resulting in permanent data loss. Brice recalled how the situation worsened and “blue screens of death” appeared on many virtual servers and key databases were lost. The IT team decided to rebuild the virtual machines and restore data and configurations from backup.
The SAN failure affected servers designated for Exchange and SQL, in turn affecting customer service. “Databarracks’s backup and disaster recovery service helped us restore IT services within a time frame that didn’t have too much of an impact on the business or our customers and suppliers,” Brice said.
WTB uses Databarracks for online backup services to safeguard its email, databases and files across 21 servers. WTB uses Asigra software for backing up its data as well as for local storage, archiving and backup lifecycle management capabilities.
As Brice made the decision to shift from WTB’s traditional in-house IT strategy, he also found that the company’s systems were very close to capacity in terms of disk space. “We were running out of resources and we decided to upgrade before we moved our IT infrastructure into a third-party data centre.”
The company’s existing IT infrastructure included different specs of hardware that caused problems such as compatibility issues. “We had a mix of servers with different generations of processors,” Brice explained.
The company had three IBM servers and five HP servers.
“We opted for three high-capacity HP servers to replace the older servers,” he said. WTB is using HP Proliant DL580G7 which features four 32 cores and 64 v CPU, and 400 GB RAM.
Advantages of colocation services
Databarracks has only hosted WTB’s full IT infrastructure and provided colocation services since the beginning of 2011. Brice and his team manage the IT infrastructure themselves, opting for just support and hosting services from the service provider.
“Our infrastructure is now held in an environment with lots of redundancy built in, something which would be too expensive for us to invest in ourselves,” Brice said. “Uptime has been improved to 99.99%, equal to 53 minutes of downtime per year, allowing more time for planned maintenance and upgrades.”
Brice admitted that the company’s in-house IT model was cheaper and had a much smaller impact on the IT budget. “But as the business grew, the priority was to have a scalable high-availability infrastructure that guaranteed business operations,” he explained.
“It also allows the IT team to sleep peacefully at night knowing our infrastructure is running in a secured data centre environment with redundant power and cooling which is constantly monitored and managed, day and night.”
Keen to save costs elsewhere, WTB decided to purchase the backup hardware and software. “Initially Databarracks provided us with an end-to-end service for disaster recovery, but we now own the hardware and software located at Databarracks’ facility and the service provider only manages it for us.”
“Our servers are replicated to our own VMware Data Recovery (VDR) hardware, which sits in the data centre,” Brice added. “If we lose servers, SANs or individual virtual machines, we can simply switch on the replicated versions using the VMware Virtual Centre and it will immediately be accessible on our network. We also have a redundant network route into the VDR environment to cover any loss of service to our MPLS WAN circuits, Internet connection or the whole of our infrastructure.”
After setting up a foolproof disaster recovery strategy and after moving away from its in-house IT arrangement, WTB is now looking to virtualise its enterprise resource planning software.
