As a substitute of 1 occasion per digital machine, now you can stack a number of situations behind a Redis proxy. There’s one other huge change: though two nodes are nonetheless used, each run a mixture of major and duplicate processes. A major makes use of extra sources than a duplicate, so this method lets you get the absolute best efficiency out of your digital machines. On the similar time, this mixture of major and duplicate nodes mechanically clusters knowledge to hurry entry and help cross-region geographic replication.
Azure Managed Redis has two totally different clustering insurance policies, OSS and Enterprise. The OSS possibility is identical as that utilized by the group version, with direct connections to particular person fragments. This works properly, with virtually linear scaling, however requires particular help in no matter consumer library you might be utilizing in your code. The choice, Enterprise, works by way of a single proxy node, simplifying connection necessities for purchasers on the expense of efficiency.
Why would you utilize Redis in an utility? In lots of circumstances, it’s a device to maintain periodically accessed knowledge cached, permitting for quick learn/write entry. Used wherever you want a quick key/worth retailer that helps fashionable options like vector indexing. Utilizing Redis as an in-memory vector index helps hold latency to a minimal in AI-based purposes. elevated restoration era (RAG). Cloud-native purposes can use Redis as a session retailer to handle state between containing purposes, so AI purposes can use Redis as a cache for current outcomes, utilizing it as semantic reminiscence in frameworks like Semantic core.
