AWS Migration Case-Study: R M Curtis

The Business 

For nearly 170 years, R M Curtis has specialised in the import and distribution of ingredients, more recently producing own-label products to a huge number of UK customers, including many familiar high street names in both the retail and wholesale markets. Having multiple infrastructures spread across multiple locations, the company started looking towards the cloud to enable standardisation across the organisation's IT estate.

The Challenge

With multiple warehouses at different locations, RM Curtis’s infrastructure was spread out – with little standardisation, limited fault tolerance, and limited disaster recovery capabilities. By working with us as an AWS Consulting Partner, we helped the business to gain a comprehensive understanding of their environment and we established the following challenges for the organisation;

• Consideration for the replication of core infrastructures such as Active Directory sites and services
• Heavy reliance on SQL Databases meant that latency and bandwidth had to be considered
• As a production facility, RM Curtis required a stringent high availability and disaster recovery plan
• Cost management and total cost of ownership are essential to keep maintained for the business, to ensure that the IT Manager could demonstrate value-for-money at a board level
• A migration plan would need to be thought out and thorough to minimize any business impact on the organisation

The Solution

By harnessing the flexible power of EC2 instances we ensured that the production servers are optimised for a balance between cost, performance, and recoverability. Once built we staged a thorough testing plan to ensure that the servers were suitable before a complete migration for all users. The migration plan was staggered which allowed the business to achieve minimal impact for the users.

Multiple VPCs were deployed – spread across multiple Availability Zones within the EU-West-1 region for different workloads, including production, test, and disaster recovery. We utilised CloudWatch and EBS snapshots to ensure that performance can be reviewed regularly.

We optimised costs through burstable, reserved instances and shutdown schedules to ensure costs were in line with the business's budget. We also utilised a comprehensive tagging strategy to ensure we can provide a detailed cost breakdown and aid automated tasks based on tag values.

Outcome and Results

Compliance and governance tracking: By enabling AWS configuration, resource configuration changes can be tracked over time. We deployed configuration rules to ensure that resource compliance can be tracked over time. By linking these rules with Event Bridge, we are kept up to date with changes to the environment in near real-time.

Increased resilience: By deploying the infrastructure across multiple availability zones we have ensured that should a critical event occur in the primary availability zone, resources deployed in the other zones can continue to work independently. This is paired with a robust snapshot schedule that is independent and includes hourly snapshots of the business's most critical data. Systems can be back online with minimum interruption to the business.

Confirmed DR capabilities: By utilization of the elastic nature of the cloud, we were able to robustly test various disaster scenarios to confirm the RPO & RTOs for key resources in the search scenario.

Test environment: R M Curtis’ core applications go through regular upgrades which must be tested prior to being rolled out to the production environment to ensure stability. This is easily reproduced at scheduled intervals, using automation and infrastructure as code.

Cost Optimisation: With a combination of reserved instances, budgets, cloud cost reviews, and AWS Trusted Advisor the company have managed to keep their costs minimal by ensuring the resources are on the right size and instance type at the right time