Skip to main content
Edge Storage leverages a distributed Redis architecture optimized for FastEdge’s global WASM runtime. Here’s what makes it unique:

Architecture

Co-located Redis instances: Unlike traditional CDN solutions that run Redis at regional Point-of-Presence (POP) locations, Edge Storage deploys a Redis instance at every edge location where your FastEdge application runs. This means your WASM component and its data store exist in the same environment, eliminating network hops and providing single-digit millisecond read latency. Write-once, replicate everywhere: When you write data through the API, it’s first persisted to a central SQL database, then automatically replicated to all edge Redis instances globally. This replication typically completes within 1-2 seconds.

Performance characteristics

  • Ultra-fast reads: Reading from Edge Storage happens locally at each edge with sub-millisecond latency since the Redis store runs alongside your WASM application
  • Eventual consistency for writes: API writes take 1-2 seconds to propagate globally, making Edge Storage ideal for read-heavy workloads

Why this approach wins

Traditional edge computing platforms place Redis at regional POP locations, forcing your application to make network calls to reach the data store. This introduces latency and potential failure points. By co-locating Redis with your WASM runtime at every edge, Edge Storage delivers:
  • Lower latency: No network round-trips between your application and data store
  • Higher reliability: Data access doesn’t depend on cross-datacenter connectivity
  • Better performance: Optimize for the 99% use case (reads) while accepting slightly slower writes
This architecture is perfect for frequently-read, infrequently-written data. Ideal use cases include configuration data, feature flags, content fragments, and blocklist/allowlist checks based on IP addresses, IP subnets, ASNs, and similar criteria.