Relocating, upgrading, and streamlining Yorkshire Housing’s entire IT infrastructure

Yorkshire Housing was formed in 2008 following the formal merger of different elements of the Yorkshire Housing Group.Today, this forward-thinking organisation manages over 18,000 affordable and social rent properties and has plans to build an additional 3,000 by the end of 2021.

A landlord that genuinely cares about its tenants and wants them to thrive, Yorkshire Housing delivers tailored services together with additional support for people with special needs to help them lead more independent lives. 

Yorkshire Housing’s single-minded goal is to provide homes and communities where people are proud to live and as a result be the first choice for anyone who wants to rent or buy a home in Yorkshire. 

THE CHALLENGE. 

Like many organisations across the world, the pandemic, ensuing lockdowns, and social isolation forced Yorkshire Housing to re-evaluate the way it was operating. As a result, it introduced a wide swathe of new working practices that included equipping its sizeable workforce to work remotely.

Consequently, with the future uppermost in mind, the decision was made to move out of its head office in Leeds city centre and establish a new, more flexible decentralised working model. The ultimate goal was to give staff more autonomy and allow them to work from smaller satellite offices, from home, even in public places such as coffee shops. 

Aside from the inevitable logistical problems of such a major relocation, it  also raised a number of major challenges with regard to Yorkshire Housing’s existing IT systems. 

First and foremost, because the head office also housed its main server infrastructure, it meant that any move would require its relocation to an alternative site. 

To add to the complexity of the project, in addition to this primary data centre which was concerned with the organisation’s day-today operations, Yorkshire Housing also has a second datacentre in a different geographic location in order to provide resilience. 

For reasons of operational continuity and security, Yorkshire Housing didn’t want to place both elements of its infrastructure in the same datacentre, as this would create a single point of failure. So, this forward-thinking organisation required some options for an alternative data centre that would be geographically separate from the current DR site. 

But that wasn’t the only challenge that Yorkshire Housing faced. 

It was also recognised that the introduction of new flexible working practices combined with future growth would soon place even greater demands on its IT systems. So, it made complete sense to use the re-evaluation and relocation of its data centres as the ideal opportunity to upgrade and enhance a number of key systems and introduce new functionality. 

Furthermore, Yorkshire Housing wanted to realise more value from the DR site as the existing setup was simply configured to receive replica data. This was viewed as being a waste of resources, not to mention money, and the organisation was keen to make greater use of its resources by also running live workloads there. 

Yorkshire Housing was also concerned about the capacity of the DR site in Manchester, which in terms of compute and storage wasn’t operating as efficiently as to its primary data centre. Yorkshire Housing wanted to increase its capacity as close as possible to 100%, bringing it in line with the efficiency of its primary site. 

Furthermore, Yorkshire Housing’s existing systems didn’t have the requisite network redundancy. Without this process that ensures network availability in the event of a network device or path failure or unavailability, there could be no automatic network failover. As a consequence, the failure of a single site could potentially take down the entire network and require manual intervention, resulting in a much longer RTO (Recovery Time Objective) to restore services. 

Faced with this raft of major challenges, that all needed to be completed in the most efficient and timely manner possible in order to minimise any operational downtime, Yorkshire Housing called upon Quadris for help. 

THE SOLUTION. 

Our first task was to identify a suitable site for the relocation of the primary datacentre. 

As requested, we explored locations that would be geographically separate. The new location had to be far away enough to avoid a single point of failure, but close enough to be logistically less challenging when it came to moving the physical elements of the primary datacentre to its new home. 

With the location agreed, we set about the task of designing a solution that would improve the efficiency, reliability, and security of the entire system. To enable Yorkshire Housing to make maximum use of both its datacentres we proposed a new enhanced solution designed around a Live-Live configuration. 

In order to implement a Live-Live configuration one of the first steps would be to bring connectivity speeds up to the mark, as the existing 500Mbps of bandwidth is only sufficient for offsite backups and replication data. Moving forward, it may be necessary to introduce two resilient 10Gbps connections between the two datacentres. This would allow for the new Live-Live configuration for all Virtual Machines (VMs) between both clusters, and with Virtual Local Area Networks (VLANs) stretched across both sites. 

With compute and storage in mind, to deliver the new levels of performance and accommodate the proposed Live-Live configuration, will require the renewal of all existing Nutanix nodes at both sites. Furthermore, the addition of 3 new Nutanix nodes to the DR site will increase the compute and storage resources and bring it in line with the capacity of the primary site. 

With regard to network redundancy, we recommended replacing the top of the rack switches and introducing new core switches at both sites using tried and tested Cisco Nexus switches. Not only will this remove the reliance on the primary site for networking should it go down, but it will also automatically failover the gateways over to the core at DR, so that the network will continue to function. 

In addition to its new-found functionality and increased performance, the proposed Live-Live configuration will also cross protect, meaning that all workloads running from the new primary datacentre will automatically replicate to the DR site and vice-versa. Not only will this deliver even greater levels of working capacity and operational efficiency, in the event of a disaster the new configuration will enjoy a very low Recovery Point Objective (RPO) and Recovery Time Objective (RTO). 

The existing Hyper Converged Infrastructure (HCI) is based on Nutanix and running on its own proprietary Acropolis Hypervisor (AHV). An excellent solution which Nutanix is continually improving with new features such as Metro Availability which will enable continuous availability across the two sites and deliver zero RPO and near-zero RTO for all critical workloads. 

Unfortunately, at the time of putting together the original proposal Metro Availability was still in development. As a result, we agreed with Yorkshire Housing that we would proceed with the relocation of the primary data centre and undertake all the preparatory work, such as implementing the networking and increasing the number of Nutanix nodes at the DR site, ready for when Metro Availability is released later in the year. At which point Yorkshire Housing can make the final decision whether to implement Metro Availability and upgrade to the necessary 10 Gbps connectivity. 

In the meantime, the system will remain as a Live-Standby configuration but will now have the capacity to replicate all VM data rather than being limited just to the more critical VMs. 

All supported by Quadris Network Managed Service. 

To help maintain both the existing and proposed infrastructure, Yorkshire Housing will also be able to call on Quadris for ongoing support. 

Should any problems arise, they can rely on our dedicated team of expert technicians combined with a range of sophisticated automated tools to deliver a wide range of operational and business benefits: 

  • Full Managed Maintenance Service. 
  • Access to Prevensys our proprietary software solution for round-the-clock monitoring of the existing Nutanix, Citrix and Commvault platforms, plus the new Cisco Nexus networking infrastructure. 
  • Unlimited upgrades outside of business hours. 
  • Network device monitoring including:
    • Network Mapping and Discovery.
    • Network Monitoring and Alerting.
    • Configuration Management (Automatic Configuration Backups).
    • Remote Management.
    • Dashboards.
    • APIs.
    • Network Flow Monitoring and Analytics.
    • Geo IP Tracking.
    • Centralized Syslog Collections.
  • Configuration changes outside of business hours.
  • Unlimited incident response and remediation 24/7/365. 

The planned implementation. 

Because of the physical upheaval of the primary site to the new location, coupled with the delay in the roll out of Metro Availability, the implementation will require a high degree of foresight and planning. 

The first phase will involve the introduction of the new Nutanix nodes and Cisco Nexus switches at the DR site. Only when we are completely satisfied that the DR site is fully functional will we turn our sights to moving the primary site.  

To minimise the downtime impact of the actual move of the primary site from Leeds to Liverpool, we will failover everything that is running on the primary site and let in run from the new upgraded DR site. 

Only when the DR site is running smoothly and problem-free, will we set about physically moving the primary site to its new home. As we will be failing over and running from the DR site the move can take place over a weekend or even a normal working week, with minimal impact on the day-to-day running of the organisation. 

Once the new Liverpool site is tested and found to be fully functional, we will move everything back from the DR site. At this point it will still be operating in the original production/DR scenario, whereby 100% of the workload is running from the new primary site and simply replicating to the DR site. 

But with all the preparatory work undertaken, when Metro Availability is released by Nutanix, we will be in the perfect position to implement the new operating system and move approximately 50% of the workload over to the DR site.

The completed project will deliver Yorkshire Housing with a new upgraded and streamlined IT infrastructure will fast, efficient DR capabilities that will meet its needs for years to come.

What challenges do you face?

To discuss your IT challenges and how we can help, just pick a slot in the calendar and Peter will be happy to help. Alternatively, feel free to call 0161 537 4980 and ask for Peter Grayson or email peter.grayson@giftalien.com 

More success stories