Payboost, a Fintech startup created in 2016 and a subsidiary of Nova Veolia, is a provider of digital payment collection solutions and the creator of the first French budget and social welfare coaching app. With three sites in France and overseas, Payboost employs 75 people serving 2,500 B2B customers and more than 500,000 users.
Handle the collection of several hundred million euros per year, leveraging the Hyper-Availability of 13 TB of data scattered across 18 host servers, 410 virtual machines (VMs) and 50 physical machines.
Payboost, a Fintech startup created in 2016 and a subsidiary of Nova Veolia, a company dedicated to innovation and the development of digital services, is a provider of digital payment collection solutions and the creator of the “Fastoche,” France’s first budget and social welfare coaching app for low-income individuals.
Payboost’s customers include the billing departments of suppliers of essential goods that have a social impact (i.e., gas, electricity, local authorities, etc.) as well as banks and individuals. This means Payboost represents a total of more than 2,500 B2B users and 500,000 B2C users that are located all over the world. In accordance with the backup requirements of the financial sector, Payboost set up an information system made up of three data centers, all with the same technical configuration: a production site, a mirror site that operates in continuity of service and a third site for disaster recovery. This means Payboost can operate without interruption to process payments totaling several hundred million euros a year.
Hyper-availability is critical for Payboost, which is committed to up to 99.99% service continuity with its customers — a responsibility that rests on an IT department that consists of only three people. Therefore, this startup needed an automation-based solution that would allow the IT team to remain agile and help free up time for value-creating projects such as new product development.
To ensure the hyper-availability of its 13 TB of data, spread over 18 host servers, 410 VMs and around 50 physical machines, Payboost chose to put Veeam at the heart of its information system’s design. “We chose to build the infrastructure with and around Veeam,” explains Eric Machabert, information systems enterprise architect at Payboost. Only Veeam guaranteed data protection that is both reliable and completely painless to Payboost’s operations and the operation of its customers and its customers’ clients. This was a prerequisite for meeting the high demands of the company’s key account customers, such as major utility suppliers.
The technical infrastructure, sizing and technologies implemented at the hardware level were chosen to get the most out of the Veeam solution and vice versa. Veeam’s native integration with Payboost’s full Hewlett Packard Enterprise (HPE) infrastructure includes 3PAR full-flash storage arrays for primary storage and HPE StoreOnce for secondary storage. This enables VMs to receive snapshots of less than one minute, as well as a total cumulative backup window of 35 minutes for 410 VMs, with data rates greater than 2 GB/second. This reduced backup window guarantees the integrity of production performance and the high quality of Payboost services.
Thanks to Veeam’s advanced replication capabilities, Payboost builds the disaster recovery environment from its mirror site every day. All the company’s production VMs are replicated from the copies of production data backups. Thus, the VMs are ready to be restarted in the event of an incident to reduce recovery time objectives (RTOs) to a minimum. Hyper-Availability also extends to the company’s 50 laptops, thanks to Veeam Agent for Microsoft Windows, which allows Payboost to restore an entire mobile computer onto a new physical machine in the event of loss, breakage or theft, for less than 30 euros ex. VAT/year per machine.
Thanks to Veeam, Payboost is also able to more accurately inform its customers of the downtime to be taken off from the maintenance times scheduled in its service level agreements (SLAs). Indeed, ahead of software deliveries or updates, Veeam allows Payboost to perform tests by restarting its production environment in isolated silos in a very simple and nearly automated way, while also taking full advantage of the performance of native production. “This allows us to plan a smooth delivery and to avoid financial penalties,” says Machabert. The rest of the time, the monitoring, reporting and capacity-planning tools in Veeam help Payboost continuously monitor and analyze all their infrastructure’s vital signs. “Thanks to the excellent visibility of our infrastructure, we immediately detect any potential problems and can act quickly,” says Machabert. Moreover, thanks to the recovery quality of the Veeam solution, Payboost feels fully confident in its ability to recover any critical file or directory that is compromised, deleted or lost by its users. “To date, we have never had any inability to restore. This gives our teams and our customers absolute peace of mind as to the Hyper-Availability of their data,” says Machabert.
In addition to the solution’s exemplary reliability, Payboost also praises its performance and costs controlled over time. “The solution we have deployed, configured and implemented delivers an optimized level of performance at the pace of updates, with total cost of ownership becoming increasingly cost-efficient over time and an overall management requiring only three hours per week from the team,” explains Machabert. “Thanks to this solution, we can focus on what truly makes our value, namely developing new products and building new solutions.”