WhiteSpider working with Liverpool University

The University of Liverpool is one of the UK’s most vibrant, diverse and forward-thinking universities. 

Founded in 1881, and a member of the prestigious Russell Group of universities, it has more than 22,000 campus-based undergraduate and postgraduates, enrolled onto some 400 different courses.

The University is currently part way through a tenyear strategic plan to further extend its reputation as a world leader in research. It also has ambitious plans to boost collaboration and inter-disciplinary working and has already established links with some 1,300 external organisations. 

The new ACI infrastructure allowed for higher throughputs and lower latency enabling all departments to access information efficiently, reliably and securely.

Agile service delivery is key

With its increasing focus on external collaboration, online learning and ground-breaking research, it is vital that the University’s underlying IT infrastructure enables agile service delivery and global connectivity.

Central to this infrastructure are the University’s two data centres, which are both based on its main campus close to Liverpool city centre. While many organisations run a primary data centre and a dedicated disaster recovery (DR) facility – which will only operate in the event of a failure or outage at the primary site – the University’s two data centres are active around the clock, with both fulfilling the role of primary and DR site on a dynamic basis.

Technology refresh

The University recognised that it needed to upgrade its data centre infrastructure. Its existing architecture – a traditional three tier infrastructure – was built on Cisco Nexus 5000 and 2000 series switches, both of which were nearing end of life. Moreover, to improve the efficiency of its data centre operations and to support a growing IT requirement, the University needed to invest in a solution that would centralise control and simplify the delivery of new applications. 

The University elected to upgrade to Cisco’s software defined data centre infrastructure, Application Centric Infrastructure (ACI). Based on a clos network topology, rather than a three-tier architecture, this would provide more reliable, scalable data centre operations. 

Following a competitive tender exercise, the University of Liverpool chose the WhiteSpider team to design, test and deploy the new data centre infrastructure. The decision was based on WhiteSpider’s unrivalled experience of managing ACI deployments in complex environments. 

This configuration was designed to enable optimal utilisation of the University’s data centre estate. 

Applying the WhiteSpider methodology

Our team followed WhiteSpider’s systematic and proven framework for project delivery, which encompasses a number of steps to ensure the solution would deliver on the expected outcomes: 

• At the outset, we undertook a comprehensive audit of the University’s existing IT infrastructure plus its various technology and business interdependencies, to ensure the new data centre would interoperate effectively from day one, as well as for the long term.
• We also hosted a workshop to explain the proposed design and rollout, providing a number of migration and deployment scenarios, taking into account the potential addition of extra virtual and physical servers, additional switches and new policies in the future.
• Once the infrastructure was fully tested and operational, the team undertook a comprehensive four-day knowledge transfer programme, to ensure the University’s network team was fully equipped to manage the infrastructure on a day-to-day basis.

Migrating to a multi-pod architecture

At the outset of the tender process, the University specified it required a significant upgrade of its data centre infrastructure, migrating from its legacy three tier architecture to a more flexible ‘clos’ architecture. As the University’s IT team was eager to manage both the data centre facilities as a single entity, the initial design concept was based on a Stretched Fabric architecture which, would enable this single point of management.

The WhiteSpider team, however, quickly identified that the emerging Multi Pod architecture would further enhance the resilience of the fabric (by providing a single availability zone split into multiple fault domains) and delivering simpler management of the active-active pair of data centres - ensuring the University’s network team could administer the entire estate efficiently as a single fabric.

Supporting new and improved applications and services

The impact of the new ACI platform has quickly been felt across the entire University, and furthermore its ambitious plans for enhancing services to staff, researchers and students are now possible with the new infrastructure.

• In the Summer of 2018, the University completed the first phase of its largest ever IT project; the roll out of a new student records system that’s used across the entire facility, including teaching departments, finance and accommodation. The new ACI infrastructure underpins this system, with the higher throughputs and lower latency enabling all departments to access information efficiently, reliably and securely.
• The University now mandates the recording of all lectures and makes them available online to international and remote students, as well as to on-campus students for revision. A low latency, highly available infrastructure is key to the success of this initiative.
• The University’s virtual learning system – Blackboard – is to be migrated onto the ACI fabric. Blackboard enables students to access learning materials, take tests and submit course work, while teaching staff can mark work, analyse for plagiarism and track student progress all on a single platform. ACI makes this project simpler and faster to achieve.
• The University’s many research programmes are also set to gain. A complete migration away from the legacy network is complete, which will play a key role in furthering the University’s already impressive reputation as world-leader for research.

The benefits have also been obvious within the data centre environment itself, with the team’s monitoring tool identifying an immediate boost in the performance of the network.