Technology & Innovation - Issue 12

“Catastrophic outages often come fromahidden supplierweakness” that reliability: “ We design for resilience from the outset, ” he says. “ Instead of one large server, we scale horizontally – adding many smaller ‘nodes’ so that the system can reroute if one fails. Auto-scaling in the cloud lets us meet unexpected spikes without wasting energy when demand is low. ” That scaling is tested constantly. “ We run load tests using tools like Gatling and JMeter, ” Ian explains. “ We simulate thousands of teachers taking the morning register, or parents rushing to top up dinner money, ramping traffic up and down to find break points. Azure’s Kubernetes technology then spins up extra computing power automatically when usage passes set thresholds. We know that every morning, more than 31,000 people will log on to take registers through our cloud version of SIMS – so it’s vital that our systems are robust enough to cope with demand. ” Sector-specific insight These planning processes also have to take account of the school calendar. “ We know that September and January bring surges, as pupils return and parents set up payments, ” Ian says. “ Our engineering and support teams work round the clock, and are laser-focused on the early morning load from 7am, so that we’re ready when logins begin and can ensure that there’s no disruption to the school day. ” That sector-specific insight is a key differentiator. “ You can buy cloud services anywhere, ” Ian notes, “ but understanding how schools operate – what time registers open, when exams are going to create extra load –means we can anticipate demand and fix issues faster. ” Finally, Ian urges decision- makers to probe suppliers’ contracts and supply chains. “ Some organisations run on a ‘minimum viable’ approach, and accept long outages because it’s cheaper,” Ian concedes. “ Building safe, reliable, scalable and performant systems is hard, and takes ongoing dedication and investment. We provide systems that can safely be relied on by tens of thousands of teachers, and also millions of school children’s parents, so we need to be able to safely rely on our suppliers. “Ask providers how they manage third-party risk, whether they operate across multiple independent processing sites, and how often they test failover. Catastrophic outages will often come from a hidden supplier weakness. ” For schools, the lesson is clear – technical resilience isn’t just about shiny features. It’s about architecture, and testing and cultivating a culture that treats uptime as mission- critical – so that the teaching and learning need never have to pause. For more information, visit parentpay.com QUESTIONS FORYOUR TECH SUPPLIER 1 How do you handle unexpected spikes? Look for horizontal scaling and automated cloud expansion that will keep systems responsive during registers, exams, or sudden parent- payment rushes. 2 What are your Recovery Point and Time Objectives? These define how much data you might lose and how quickly service will be restored after a failure. 3 Do you run active- active disaster recovery? True high availability requires two fully capable sites, so that the service can switch over instantly if one goes down. 4 What independent certifications do you hold? ISO 9001 (quality), ISO 27001 (security), and PCI DSS indicate mature processes and regular external audits. 5 How do you manage supply-chain risk? Ask about the financial health of key partners and whether network routes and data- centre providers are fully diversified. 6 Will your support hours match school reality? Early-morning coverage is essential when staff and parents log in before lessons begin. By pressing for clear, evidence-based answers, schools can choose partners who will keep their learning on track – even when the unexpected happens. 55 A D M I N I S T R AT I O N teachwire.net