All Customers / Customer Story
eBay Inc. (NASDAQ: EBAY) is a global commerce leader including the Marketplace, StubHub and Classifieds platforms. Collectively, they connect millions of buyers and sellers around the world, empowering people and creating opportunity through Connected Commerce. Founded in 1995 in San Jose, Calif., eBay is one of the world’s largest and most vibrant marketplaces for discovering great value and unique selection. In 2016, eBay enabled $84 billion of gross merchandise volume.
InfluxDB is used by various teams at eBay for DevOps monitoring and real-time analytics - here are some testimonials from different teams.
Elasticsearch as a Service at eBay
eBay has built an internal Elasticsearch as a service (ES-AAS), a fully managed Elasticsearch cluster to support the various search use cases found internally within eBay. This project involves deployment and configuration of an Elasticsearch instance on an eBay OpenStack internal cloud system. This system uses InfluxDB to store health metrics of this ES-AAS at a rate of about 2,000 metrics per minute as well as use Grafana for dashboarding. This InfluxDB/Grafana combination gives the team quick insight into the health of their ElasticSearch cluster and helps them take the proper remediation action.
eBay's Experimentation Team
InfluxData’s InfluxDB and Grafana are used by eBay’s experimentation platform to consistently measure and monitor various data quality metrics. eBay’s Experimentation environment enables eBay’s business users to answer important analytics and business questions. This testing allows business units to conceptually explore new ideas that can vary from subtle to radical variations. Such test variations can be easily targeted to segments of the total customer population providing a level of assurance before launching to a broader audience.
- Blog: Monitoring Anomalies in the Experimentation Platform
- Blog: Scalable and Nimble Continuous Integration for Hadoop Projects
Anomaly & traffic prediction
eBay is in the business of detecting various anomalies in their data stream and use different machine learning techniques and algorithms to predict normal and strange traffic patterns associated with their traffic and data stream. They use InfluxDB to store the results of their anomaly/traffic prediction which runs close to a billion records. Having this analyzed data represented in time series format is key and helps them to present it in their Grafana dashboard. They chose InfluxDB due to its scalable nature which makes the storage and retrieval of time series a cinch. They also appreciate the simple setup and responsive APIs which make managing their solution easy.