ADEO Services: Equipping retailers with accountability and monitoring in a hybrid cloud context
About ADEO Services
ADEO’s mission is to inspire homeowners and help them create their dream home. Supporting this mission is the data platform team at ADEO Services, which helps ADEO make the most of its data, respecting privacy, security, availability, durability, consistency, and performance, so that it can serve its 452 million customers more effectively.
Tell us your challenge. We're here to help.
Contact usADEO Services’ data platform SRE team used the Google Cloud operations suite to build tools that enable data platform engineers to focus on data-driven initiatives that help ADEO meet customers’ needs.
Google Cloud results
- Speeds time-to-action when software issues arise by storing cloud and on-premises logs in one integrated interface
- Makes it easy for site reliability engineers to improve the experience of users with automated SLO monitoring
- Increases performance, uptime, and health of applications with Cloud Monitoring, which automates alerts of emergent issues
- Consolidated accountability database into BigQuery to store around 650 TB of logs long-term and improve analysis across lines
All logs across environments ingested in real time
There’s no place like home, and ADEO wants to make it possible for everyone to create the home of their dreams. Together, ADEO’s different companies have 800 points of sale in 15 countries, including market-leading retail outlets, warehouse stores, and innovative concept stores. Something else they have in common? ADEO’s very own data platform team, whose business purpose is to collect, store, and deliver the capabilities that enable all of ADEO’s companies to search, consult, and use data easily.
The data platform was established in 2018, when ADEO took on a digital transformation project. The data platform team adopted a site reliability engineering (SRE) model to administer the platform, meaning that a dedicated team was put in place to focus on keeping services running and users happy while identifying opportunities to automate repetitive work. To operate in a systematic way, at scale, while staying secure and compliant with company policies, the team uses the Google Cloud operations suite to monitor services using self-managed solutions.
Around the same time, the team worked with Cloud Consulting Services to co-develop a serverless tool known as SLO Generator for use among the open source community. This tool uses Cloud Monitoring as a metric back end to compute and export critical service SLOs, error budget, and burn rates based on configuration files.
“The data platform team’s mission is to help create additional business value for the wider organization through data and innovation. As part of our team, SREs support its most critical digital services. That’s what adopting an SRE mentality means to us, and Google Cloud empowers us to do it.”
—Sébastien De Nef, Site Reliability Engineer, ADEO ServicesThese are just two examples of how the ADEO Services data platform team helps this division to deliver valuable data across ADEO’s 32 global subsidiaries, enabling them to serve their customers more effectively. With the right data to inform their proposals, ADEO Services creates initiatives using AI, for example, to predict which products or services will be of most interest to a customer, based on their online interactions and preferences.
“The data platform team’s mission is to help create additional business value for the wider organization through data and innovation,” explains Sébastien De Nef, Site Reliability Engineer at ADEO Services. “As part of our team, SREs support its most critical digital services. That’s what adopting an SRE mentality means to us, and Google Cloud empowers us to do it.
Automating key services to help users stay productive
One of the SRE team’s most important functions is to alert the data platform teams if critical service users experience any problems. For this, it relies on solutions in the Google Cloud operations suite, such as Cloud Monitoring which enables the teams to gain visibility into the performance, uptime, and overall health of their services.
Beyond this, however, the SRE team’s main purpose is to develop automated solutions for processes that can be streamlined and improved, and one way to do this is to build self-service tools.
Cloud Monitoring gathers metrics, events, and metadata from all the services on the data platform, and generates insights that the data platform SRE team can use to identify potential issues. To implement this solution, the data platform SRE team at ADEO Services worked closely with Cloud Consulting Services to define the appropriate metrics to be measured in its system, based on what its users need. Together, they defined the data platform service level objectives (SLOs), which help to indicate whether all services are healthy, and recommend the right action if something goes wrong. Relying on this one integrated service for metrics, dashboards, uptime monitoring, and SLO alerting means that the team can spend less time maintaining disparate systems and more time delivering value for users.
“Our SLO framework helps us gauge how our services are performing and what can improve for end users based on metrics and logs that can be very time consuming and challenging for us to find. With Cloud Monitoring automatically collecting them for us, we can focus on actioning the resulting insights to improve services for users,” Sébastien explains. This SLO framework is based on Google Cloud operations tools and completed with the SLO Generator to manage SLOs easily and at scale, and integrates PagerDuty as the paging system for on-call users.
“Logs are crucial for helping technical teams understand what’s happening in a software. Storing them in the same back end allows us to plug multiple data visualization tools according to who is consuming. We don’t have to look at disparate systems to find the origin of the problem, so we can fix it faster.”
—Sébastien De Nef, Site Reliability Engineer, ADEO ServicesGaining speed to determine the origin of production issues
Meanwhile, the ADEO data platform SRE team also uses Cloud Logging as its universal back end for logs across Google Cloud services, on-premises, and SaaS services. This is in accordance with ADEO’s policy that all logs are automatically activated on every project in the organization. The solution stores, searches, analyzes, monitors, and alerts the data platform team on log data and events from Google Cloud and other sources, such as virtual machines (VMs) hosted on-premises. This fully managed service can ingest application and system log data from thousands of VMs and analyze them in real time. The team also created an accountability data lake based on log sinks to BigQuery ML for use in security and compliance. This accountability data lake store logs on a long-term basis and improves transversal analysis capabilities. Then they are segregated for each scope, mostly per business unit, into dedicated BigQuery datasets. Managing permissions to access these datasets can be done autonomously by each data owner through the Data Finder, a homemade application.
Previously, to collect log events from different systems, the data platform SRE team needed to know what VMs to connect with manually. If the virtual machine was down, software engineers wouldn’t be able to access the logs, so they wouldn’t be able to fix the issue. “For our SREs, this was a very frustrating problem,” Sébastien explains. “It was like hiring a mechanic to fix a problem with your car, but forgetting where you parked it.”
“The Google Cloud teams are very helpful and attentive to our needs. Together, we’re building more self-serve features to enable not just technical teams but also colleagues in more diverse roles to adopt our SRE mentality so that we can become more innovative as a global network.”
—Sébastien De Nef, Site Reliability Engineer, ADEO ServicesUsing Cloud Logging as a universal back end allows them to provide these logs through the right data visualization tools according to customers’ expectations and requirements. “Logs are crucial for helping technical teams understand what’s happening in a software. Storing them in the same back end allows us to plug multiple data visualization tools according to who is consuming. We don’t have to look at disparate systems to find the origin of the problem, so we can fix it faster,” explains Sébastien.
The SREs working on ADEO Services’ data platform team believe that these innovations are empowering all business units with smoother, more agile services. Next, Sébastien hopes to use the same kind of innovative thinking to empower other employees across functions to self-serve when it comes to making their work more data driven. He sees the Cloud Consulting Services team playing a key role in this mindset adoption, as well as the Google Cloud operations product team. “The Google Cloud teams are very helpful and attentive to our needs,” he says. “Together, we’re building more self-serve features to enable not just technical teams but also colleagues in more diverse roles to adopt our SRE mentality so that we can become more innovative as a global network.”
Tell us your challenge. We're here to help.
Contact usAbout ADEO Services
ADEO’s mission is to inspire homeowners and help them create their dream home. Supporting this mission is the data platform team at ADEO Services, which helps ADEO make the most of its data, respecting privacy, security, availability, durability, consistency, and performance, so that it can serve its 452 million customers more effectively.