This week we continue with our short stories told by those who are at the very heart of ReGen. In this post Florent, ReGen’s team leader, tells us about service delivery and the challenges of keeping an application online 24/7, year after year.
Florent has a background in computer networks and distributed systems, and has a Masters Degree in Computer Sciences. He has been involved with ReGen since its inception roughly 6 years ago and has played a core role in 'making it happen' at each stage of the project.
Florent: 'I have been involved in this project from the start. I find it very interesting to bring together many aspects of soil science combined with near real-time data into a software tool to improve productivity and sustainability for our customers.
At ReGen, I have two complementary roles with solution architecture and service delivery.
With solution architecture, we decide on how to shape the different components of the application to satisfy our business and product objectives. This is particularly important because so much of ReGen’s services rely on automated functions, primarily around data processing and modelling. Like in a puzzle, all the components of the application need to fit and interact perfectly.
Service delivery relates to all aspects of the infrastructure, services and processes that need to be put in place to ensure the application is always online. This objective requires clear focus on best practice and consideration in all activities undertaken around the application. Starting with how we host the ReGen application online - from the beginning of this project, we opted to host in a high availability cloud solution. This means our hosting provider guarantees that the environment in which our application lives will not be unavailable for more than 5 minutes per year.
We also put in place a number of processes to support our service availability aspirations. This includes continuous system backup, managed firewall, 24/7 escalation processes (if there was a problem, the issue would be progressed quickly), three separate monitoring systems and many others. As an example, our main database is backed up seven times daily with four backups kept offsite for disaster recovery.
We have also put a strategy of continuous improvement in place by which we analyse, record and mitigate any issue that may impact or has impacted the service. The ReGen service relies on challenging data acquisition from sensors located on customer site and through third parties. In order to mitigate this, we have put in place an automated assessment of data suitability. It allows us to know before our customers when a problem may start impacting their service.
We are happy not to have encountered any unplanned outage since our launch in March 2010. We are continuing to work on improving our service reliability and to always provide more sustainable farming services to you.'