Like other UK Government agencies, Forest Research must deliver highly functional, secure and cost-efficient online services. One way to achieve this is to migrate costly on-premises applications to the cloud – and that’s exactly what Rowe IT has done, on time and budget, with Forest Research’s business-critical Tree Alert system.
Forest Research is Great Britain’s principal forestry and tree-related research organisation. Born out of the Forestry Commission’s devolution, Forest Research is now internationally recognised for its evidence and scientific services – working in cooperation with stakeholders from Government departments and land owners, the European Union, and various nation states to develop British forestry policies and foster sustainable forestry management.
The challenge: A tight deadline to migrate Tree Alert to the cloud
When the Forestry Commission was split into separate divisions across England, Scotland, and Wales, responsibility for systems, applications, and data needed to be urgently reassigned to newly formed agencies.
But this posed a significant problem: all forestry and tree-related data resided in on-premises systems owned by the Forestry Commission Information Service, which because of the devolution process, was due to be disbanded and its data migrated.
One such application was Tree Alert – Forest Research’s tree pest and disease reporting portal. It allows people to upload detailed descriptions of diseased trees, with location information and photographs. This information is then evaluated by Forest Research staff who can dispatch field workers to conduct further investigations if necessary.
The Forest Research team found the on-premises database and server and storage infrastructure used by Tree Alert to be unfit for purpose, due to its slow performance and substandard, time-consuming reporting.
More importantly, Tree Alert is used to process personal and identifiable information, but the system was out-of-date and wasn’t as secure as it could have been.
Forest Research didn’t just want to migrate Tree Alert system to a secure and reliable solution in the cloud. It also wanted a reliable partner to manage and support the system to ensure security and performance is retained.
The solution: a reliable, secure new home in the AWS cloud
After conducting a search on the G-Cloud digital marketplace, Forest Research selected Rowe IT for the migration work. As an early exercise, each organisation drew up a list of problems with the existing system, with their ideas for changes that could be made.
Jon Nowicki, Project Manager for Forest Research, explains: “In one of our initial meetings with Rowe IT, both parties came to the table with lists of what they thought was wrong with the system and their proposed solutions. To our surprise, Rowe IT’s list was almost identical to ours. It helped completely cement our working relationship from the off”.
Forest Research had two simple requirements for Tree Alert: it had to be reliable and it had to be secure. Based on previous migration projects, Rowe IT recommended a move to the Amazon Web Services cloud – with an architecture based on its EC2 compute environment EFS elastic file storage service, SQS messaging service, SNS notification service, and RDS relational database service
Following a planned migration to a test and development environment, the final migration took place just three months after the project kicked off. And while both teams planned for four days of downtime during the migration process, Rowe IT completed the process in three.
“Because of the department’s imminent restructure, any delay to this project was simply unacceptable. But without sufficient communication and collaboration between stakeholders, projects can very easily fall behind schedule. Thanks to Rowe IT’s experience and resolve, we were never in any doubt we’d hit our deadline. I never lost any sleep thinking about it.”
Jon Nowicki, Project Manager, Forest Research
Tree Alert also needed to be able to pass data to a separate system – Forest Research’s on-premises Tree Health and Diagnostics Oracle database. Rowe IT re-engineered Tree Alert’s Java-based back-end extract, transform and load (ETL) process, and securely integrated it with the Oracle system for seamless data transfer.
The Rowe IT team also re-engineered some of Tree Alert’s web interface, replacing its proprietary GIS-based mapping system with a simpler open-source drop-pin map. This not only removed licensing costs, but also greatly improved the user experience.
The results: A fast, efficient application that frees up more time for research
With Tree Alert running securely in the AWS cloud, Forest Research can now process data and respond to notifications far faster – enabling staff to spend more time on more important aspects of their jobs, such as vital research into the causes of tree disease.
What’s more, better workflows and enhanced visibility over data mean the team can review information and pass it on to the right people or departments faster and more easily.
Forest Research also now finds that more of its European colleagues are aware of the system, and regularly add to its growing resource of data and trend analysis to better predict and prevent the spread of disease in trees.
The team continues to work closely with Rowe IT, who provide continuous monitoring, support and maintenance for Tree Alert as a managed service. With renewed confidence in the system, senior Forest Research personnel have relayed its success to partner agencies and organisations such as Defra and the Woodland Trust.
Forest Research mostly relies on external funds, so before expanding any of its programmes, it needs to demonstrate tangible results. The project’s success now means it has the go-ahead to investigate the feasibility of expanding its Tree Alert system to provide an even better service.
“As we had a fixed migration deadline that couldn’t be extended even if something went wrong, it was vitally important we worked with a vendor that we completely trusted. With Rowe IT we got this, and so much more. It was an extremely slick process and the team worked closely with us to ensure all possible risks were mitigated.”
Jon Nowicki, Project Manager, Forest Research