The Universities and Colleges Admissions Service (Ucas) assigned university places to almost 400,000 UK students on 14 August.
The IT platform it uses, which feels the strain for a few hours on a single day in August, used both Amazon Web Services (AWS) and Microsoft cloud services to dramatically scale up the infrastructure for a few hours to meet the spike in demand.
By submitting your email address, you agree to receive emails regarding relevant topic offers from TechTarget and its partners. You can withdraw your consent at any time. Contact TechTarget at 275 Grove Street, Newton, MA.
Ucas, the organisation that matches students to universities, processes in excess of 2.5 million applications every year for 700,000 prospective students across the UK, and helps them gain access to more than 350 UK universities and colleges, making it, for one day in August, one of the most accessed websites in the UK.
After using Microsoft and AWS cloud services for the first time in 2013 for A-level results, Ucas turned to cloud services once again this year.
Last month, Ucas’s IT team performed around 700 million test transactions on AWS before A-level results day, in readiness for the placement of 400,000 students into higher education – supporting more than 230 logins per second and 350,000 course searches.
Putting critical IT infrastructure in the public cloud
In addition, Ucas hosted much of its critical back-end infrastructure in the public cloud for results day. It used Amazon services including EC2, S3, Amazon EBS, Elastic Load Balancing, Amazon Virtual Private Cloud, AWS CloudFormation, Amazon CloudWatch, Auto Scaling and Amazon Relational Database Service.
More on Amazon Web Services
These services allowed Ucas to dramatically scale up its infrastructure, giving universities a robust technology platform from which to efficiently place UK students.
Cloud services allowed the organisation to scale up and meet the demands of students and universities without investing in expensive hardware that would sit idle until results day next year.
Cloud computing technology is core to Ucas’s strategy to continually improve the university admissions experience, ensure the fairness of the Clearing system and make the process as stress-free for students and as efficient for universities as possible, according to James Munson, director of IT at Ucas.
Using the AWS cloud, Ucas was able to create an IT infrastructure that could rival most FTSE 100 enterprises for the 10 days prior to results day, he added.
“AWS ran a key part of our back-end system, which gave us flexibility and almost unlimited capacity to meet the extreme demands at the peak of the admissions cycle,” said Munson.
Cloud boosts capacity for one day per year
Meanwhile, Microsoft Azure provided Ucas with a way to scale its compute power to meet the huge peak in users, which occurs for just a few hours each year on results day.
More on Microsoft Azure
Ucas also worked with Microsoft to provide a rich monitoring service at both the application and platform levels, allowing the organisation to dig deep into its live web application’s performance and help it diagnose any performance issues.
“With the number of applicants continuing to grow, we are investing for the future to ensure Ucas delivers the best service to the students and universities of the UK. Both students and universities expect a high level of service with their admissions process, and cloud computing is a technology that we rely on,” said Munson.
If Ucas had not adopted public cloud, supporting a massive influx of traffic for just one day a year would require investment in a large amount of technology hardware, which would be unfeasible for a non-profit organisation.
Ucas also deploys Splunk Enterprise across 40 servers and about 70 log sources through AWS, and everything is forwarded to a Splunk server for indexing.
By indexing, searching, alerting and reporting on machine data from data sources across the IT infrastructure, Splunk gives the IT team a series of visualisations of its system performance, key operational metrics, usage, the queries it is running and how the various applications are functioning.