Facebook·hard

Design Typeahead Suggestion/Autocomplete

Design a typeahead suggestion/autocomplete system that provides relevant search results as users type, which is challenging due to the need for low-latency responses and handling a large volume of concurrent requests. The system should be able to handle millions of users and provide accurate suggestions within milliseconds.

Functional

▸Handle user input and provide relevant search suggestions
▸Support multiple data sources for suggestions, such as user profiles and product catalogs
▸Implement a ranking algorithm to prioritize suggestions based on relevance and user behavior
▸Handle errors and exceptions, such as network failures and data source unavailability
▸Support multiple languages and character sets

Non-Functional

▸Handle 10,000 concurrent requests per second with an average latency of 50ms
▸Achieve an availability SLA of 99.99% and a uptime of 23:59:59 per day
▸Support 100 million registered users with an average of 10 searches per user per day
▸Store and index 1 billion searchable records with an average size of 1KB per record
▸Handle a peak traffic of 50,000 requests per second during holiday seasons

Add at least 2 components to run