Neither is better in general, because they were built for different jobs. Redis is a general-purpose in-memory data store: it holds strings, hashes, lists, sets and streams, supports pub/sub and optional persistence, and is used as a cache, a session store, a queue and a leaderboard, with clients in almost every language. ScaleOut StateServer is an in-memory data grid (IMDG) aimed at .NET and Java applications. Its own product pages describe it as a distributed cache that spreads serialized objects across a cluster and adds servers without reconfiguring the application, with patented quorum technology for handling server failures. The Pro edition adds a compute engine that runs parallel analysis on the data where it sits. Its docs say it runs on Windows and Linux and that Java and C# clients can share the same data. So if you're a .NET or Java shop that wants object caching and session state that scales across a server farm, ScaleOut is built for exactly that. If you need a language-neutral store, pub/sub, or a cache shared by many kinds of services, Redis is the usual choice. For a real decision, test both with your own object sizes, read/write mix and failure scenarios, and check current licensing and support terms, since those change.
Also discussed here:
- What is the open source software Redis?
- What happened to ScaleOut Software for servers?