TalksAWS re:Invent 2025 - Learn How Sky Powers the Future of Streaming with Redis Cloud (API209)
AWS re:Invent 2025 - Learn How Sky Powers the Future of Streaming with Redis Cloud (API209)
Scaling Streaming with Redis Cloud: Sky's Transformation
Introduction to Redis
Redis is the most popular in-memory database and cache, with over 10 billion Docker pulls
Redis was recently voted the most loved non-relational database and the most used data management tool for generative AI
Redis 8.4.4 is the fastest Redis version yet, with up to 90% latency improvement and better memory compression
Redis Cloud Overview
Redis Cloud is the most reliable, available, secure, and cost-effective way to scale Redis in any cloud environment
Redis Cloud Essentials is a fully serverless, low-cost platform for smaller-scale applications
Redis Cloud Pro offers high-scale, high-availability, and high-performance Redis, with dedicated clusters and multi-tenancy
Recent Redis Cloud Innovations
Redis Flex allows using both RAM and SSD to scale databases up to 50TB at up to 75% cost savings
Faster scaling with 40% faster scale up/down and upcoming atomic slot migration
Smart client handoffs to eliminate disruption during planned events like scaling and upgrades
Better performance with AWS Graviton-based instances
Redis for Generative AI
Redis is a key building block for generative AI applications, providing fast in-memory storage, semantic search, and integrations with top AI frameworks
Redis Lane Cache provides semantic caching for generative AI models, delivering up to 15x faster performance and 90% cost savings
Sky's Streaming Transformation with Redis Cloud
Bookmarking Challenges
Sky's global streaming platform had millions of bookmarks, causing scaling and cost issues with Cassandra
The architecture was unable to meet requirements like expanding bookmark retention
Redis Cloud Solution
Evaluated Redis open source but chose Redis Cloud to eliminate operational overhead
Leveraged Redis Cloud's autoscaling, cross-region replication, and data structures like strings and sorted sets
Implemented a hybrid architecture with Redis Cloud as a caching layer and AWS Keyspaces for long-term storage
Achieved 20x reduction in write pressure, enabling migration to Keyspaces and significant cost savings
Additional Use Cases
Concurrency management: Using Redis Cloud to track and enforce concurrent streams across devices
Generative AI (Geneap): Building an in-house generative AI platform on top of Redis Cloud and AWS services
Key Takeaways
Redis Cloud provided the scalability, performance, and cost-effectiveness to transform Sky's streaming platform
The hybrid architecture with Redis Cloud as a caching layer significantly reduced write pressure and enabled migration to a managed database
Redis Cloud's features like autoscaling, cross-region replication, and data structures were critical to the success of the project
Sky is expanding the use of Redis Cloud to power additional use cases like concurrency management and generative AI
These cookies are used to collect information about how you interact with this website and allow us to remember you. We use this information to improve and customize your browsing experience, as well as for analytics.
If you decline, your information won’t be tracked when you visit this website. A single cookie will be used in your browser to remember your preference.