SmugMug is a premium online photo and video sharing service with millions of users storing billions of personal photos and videos. The company operates on a subscription-based model that offers its customers with unlimited photo uploads, safe back-ups, easy sharing, customization, image protection, and privacy features. Using rich metadata generated by photos and videos, customers can search and browse through their content to find what they care about.
The company built its own search infrastructure using open-source, database technologies.
Despite using fast on-premise SSD servers and employing engineering best practices, search query times fell as the number of users and amount of content grew.
Needed to find a highly available, scalable, and cost-effective search solution that would deliver a good customer experience.
Moved its photo, album, and user search service onto Amazon CloudSearch, migrating years of data and integrating with the simple, REST API for search and display features.
Designed a solution that could handle user-growth by leveraging Amazon CloudSearch’s auto-scaling feature.
Designed to deliver consistent low-latency (median below 50 milliseconds) and highly relevant search results to customers, while handling tens of millions of index updates per day.
Designed to handle 6 TB of rich metadata generated by user uploads over 2 year.
SmugMug saves more than $300,000 per year in operations and maintenance costs.
No longer needs to manage a search backend and can now focus valuable developer resources on developing a great photo sharing service.
“Amazon CloudSearch allows us to focus on our product by providing a high-performance search back-end for our rapidly growing customer base.”
Don MacAskill, CEO & Chief Geek