DataArt Helps James & James Modernize Their Industry Leading Fulfillment Platform

Nombre de la compañía

Ubicación

Reino Unido

Industria

Client

In 2010, James Hyde and James Strachan launched their business when they realized that the traditional technology behind order fulfillment was not good enough. They were attempting to sell honey online and encountered multiple poor and uncoordinated systems. They formed James & James (J&J) to smooth out this process for sellers.

With warehouses across UK, Europe and North America, J&J offers a high-quality order fulfillment platform for fast-moving consumer goods in multiple categories including fashion, cosmetics, personal accessories, gifts, organic products, food and supplements, and luxury goods, among others. Unlike traditional logistics providers, J&J acts as a single point of contact for product receiving, picking, packing, storage, inventory management, customer support and shipping. The J&J platform integrates with multiple e-commerce platforms (e.g. Amazon, eBay, Shopify, WooCommerce, etc.) as well as offering a suite of APIs allowing direct integration where preferred. This ensures timely order processing and notifications throughout the fulfillment journey.

Background

Initially, J&J hosted their fulfillment application on a managed service provider platform. However, their experience did not offer the scale, availability and durability needed as the business grew. Ultimately, this prompted the change to a more robust cloud solution that could scale and perform alongside their increasing needs. J&J selected AWS for these reasons.

Migration to AWS started with the database being moved to RDS, followed by compute nodes, DNS changes and gradually decommissioning the old service provider. A multi-AZ configuration allowed J&J to achieve their availability targets and improve customer experience.

Challenge

With the infrastructure foundations now shored up, it was time to extend the system and make it more flexible. The customer’s technical leadership chose to adopt a service-oriented architecture, underpinned by microservices. The other objectives were to improve the DevOps automation (especially Continuous Integration) and to enhance the cloud infrastructure provisioning and management practices through Infrastructure as Code (IaC).

J&J elected to engage DataArt as a reputable AWS consulting partner and a specialist in designing and implementing custom solutions for the retail industry.

Partner solution

One of the first items DataArt assisted with was to build the first microservices and then deploy them to Kubernetes via Amazon Elastic Kubernetes Service (EKS). The original Linux EC2-based workloads were also transformed into containers and deployed on EKS for the purpose of operational consistency.

In addition, multiple enhancements were implemented as smaller applications were built, deployed and updated with relative speed and interfaced with the monolithic system. One of the key methods for building these apps is the Functional Web App pattern, which incorporates the use of serverless functions.

From the IaC perspective, the infrastructure deployed into AWS was a mix of console "click-ops" and Terraform scripts residing in GitHub. As a starting point, provisioning cloud resources manually by clicking through menus was acceptable; however, the customer’s vision always was to move towards properly documented, versioned, and auditable states of their infrastructure. Over time, the DataArt DevOps team were able to help increase Terraform IaC automation coverage to 95%.

In parallel, J&J and DataArt identified a roadmap of CI/CD improvements. DataArt DevOps engineers focused on the CI buildout leveraging Jenkins, implementing over 400 integration tests in the process as well.

As part of the overall effort to enhance workload resiliency, Amazon Managed Streaming for Apache Kafka (MSK) was added to the solution. Introducing Kafka into the architecture was a game changer that added flexibility and reliability to the messaging and streaming infrastructure.

Besides EKS, RDS, Route 53, MSK and DMS (Data Migration Service), the solution also leverages Lambda and API Gateway. S3, Athena and QuickSight are used for BI reporting, CloudWatch for systems monitoring, and DataDog for Application Performance Monitoring to measure the user experience.

Outcomes

As a result of J&J’s collaboration with DataArt, the customer was able to implement and roll out multiple improvements, including:

Important new and enhanced integrations, such as a tax and duty (Global-e), shipping (DPD, DHL, ParcelForce) and client retail ERP system integration (BrightPearl)
A new feature for inventory putaway, which designated certain physical areas of the warehouse to a specific client
A simplified process of aggregating and managing trolleys at Inbound
A new picking system which divides picking trolleys into separate logical bins. This allows orders with more than one item to be grouped together, simplifying the follow-on pack process.
An overhaul to picking application that allows users to more easily navigate their picking route and the shelf the item has been placed on, reducing pick time and increasing picker efficiency.
Code quality and IT operational improvements through better CI/CD and greater IaC coverage

Contáctanos

Por favor, deja tus datos y te contactaremos a la brevedad.

Nombre*

Apellido*

Correo electrónico*

Rol

Empresa*

Teléfono

¿Sobre qué te gustaría conversar?*

¿Cómo te enteraste de DataArt?

¿Cuál es tu función/rol?

Acepto recibir el boletín mensual de DataArt.

He leído y acepto la Política de Privacidad*