Case Study

Platform9 empowers global food retailer accelerate store digitization by two years

Case Study: Global Coffee Retailer

Summary

Icon: Automated Deployment

Full end-to-end GitOps automation enables app deployment in less than an hour

Icon: Cost Reduction

80% reduction of per-store hardware + software cost

Icon: Central Management

The retailer chain now centrally manages 2000+ US stores

About the Retailer

The global food retailer operates 30K+ stores across the globe.

  • As a digital-first company, they build innovative in-store applications to deliver delightful customer experiences.
  • Their innovative order processing application enables customers to order personalized products via mobile/online and have it delivered as soon as they walk into the store.
  • Their Digital Video Recorder (DVR) app running in the store is used for security surveillance.
  • They are also rolling out new containerized apps for playing music in the stores.

The Challenge

The retailer was facing a number of significant challenges in running state-of-the-art apps across its 1000s of stores. Each store location ran its own siloed hardware with up to three servers.

  • The company first ships servers to retail stores and then sends installation crews onsite. This becomes a mammoth and costly task with thousands of stores to serve, including constant updates and maintenance.
  • They attempted for at least two years to develop their own distributed store management solution with VMware, Red Hat and open source software. But they were unsuccessful in operationalizing the in-store infrastructure at scale.
  • It was technically very challenging to develop a single unified and shared platform for consistent use by multiple users and domains such as operational teams, developers, and technicians.
  • Their core in-store apps were not virtualized. More apps meant more servers, adding to the cost.
  • With 1000s of outlets, the company found it challenging to create lean server-per-store and admin-per-store ratios within small facilities.

The Solution

Platform9 established an effective method to get the store up and running in hours while also giving them the power to remotely update and enhance their in-store applications in real-time.

Platform9’s unique SaaS architecture and cloud operations model empowers the retailer to deploy and manage distributed infrastructure sites remotely without requiring onsite visits by technicians for troubleshooting and support.

Platform9 enabled the retailer to maximize utilization of their in-store servers. And equipped them with a plug-and-play solution to remotely deploy the apps through a SaaS control pane.

Our SaaS management plane orchestrates the delivery of various capabilities such as containers (to be deployed soon), hypervisors, storage backends, and network backends.

They no longer have to worry about the operational burden of uptime/ SLA management, upgrades, security patches, and production outages.

The retailer deployed regional and WAN retail architecture. Platform9’s centralized management view enables global control of all regions and corresponding stores with a single-pane-of-glass approach.

Each store deployed a similar architectures to foster networking consistency, lower costs, and smaller IT footprints. A small server (32 GB RAM, 4 cores, 1 TB storage) can serve as a clearinghouse for transactions from all POS terminals, security cameras, and other devices.

The in-store app is an intelligent order management solution that accepts and schedules customer food and beverage orders (including mobile orders). It further balances available ovens, machines, and terminal loads with customer arrivals. It also controls waiting times to increase customer satisfaction and speed of service.

Diagram: Platform9 Single Pane Approach

Each store connects to the main-office network via an in-store WAN router to the VPN internet connection and then to a managed Kubernetes instance at their headquarters. Each location typically includes a router with different networking policies, CCTV and POS/IoT devices, and customer access to the hotspot.

At the store, someone can take a generic server off the shelf and plug in the network port. The new server bootstraps itself into an OS running Kubernetes. It then connects to clusters in an AWS environment that provision it with all the latest apps. In less than 30 minutes, the store is back online.

Platform9 monitors all infrastructure 24X7 with a 99.9% uptime SLA. If a physical component such as a server fails, it alerts the regional or main office’s IT group.

Platform9 Infrastructure

The Benefits

The retailer didn’t need a virtualization management solution. They needed to manage their store infrastructure and applications with SaaS. They needed easy, they needed scale, they needed fast and agile. They needed the best of cloud, designed for thousands of stores distributed worldwide.

By deploying Platform9 and operationalizing this approach, the retailer achieved their goals of scale and cost effectiveness:

  • Quick delivery of innovative applications to address fast changing customer needs and desires.
  • Fewer expensive hardware investments and refreshes by using existing footprint to rollout new applications.
  • Reduced operational costs with automated and centralized remote management capabilities.

Want to create your own software-defined store?

With the advanced Kubernetes and edge computing, it is possible to deploy software-defined stores that deliver the agility and ease of use similar to a public cloud.

Read blog

Learn how Platform9 remotely automates the deployment of all layers of the infrastructure stack in a store: Bare Metal, Virtualization, Kubernetes, Networking and more.

Watch demo

Learn the innovative use cases describing how to solve challenges of managing 1000s of retail stores with Kubernetes, creating a store of the future with SaaS, and more.

Read whitepaper

The browser you are using is outdated. For the best experience please download or update your browser to one of the following:

GigaOM’s Radar report for Managed KubernetesRead Now