Hiperspace is an acronym for “high performance space”, it provides the simplicity of accessing data as if in memory with higher performance than a conventional database. An element is immutable once bound to a Hiperspace, but can be versioned. Versioning avoids most of the scenarios where a lock manager would be required.

Hiperspace is like virtual memory but instead of each address being 32, 64, or 128 bits long, it is variable - they serialized bytes of the key are the address. When the total size of addressable memory is significantly larger than the physical memory of the computer, three advantages become significant:

• The entire space does not need to be loaded before any objects can be referenced - like Intel Optane persistent memory it is addressable immediately
• The space does not become fragmented when new objects are allocated, resulting in a large number of page faults when objects need to be loaded to physical memory
• It does not cause the computer operating system to crash when physical memory is exhausted and unable to handle page-faults due to the number of concurrent threads.  

Hiperspace is like a database accessed with "Object Relational Mapping" (ORM) and lazy loading, but without the translation of a reference to a database query using SQL, resulting in higher performance. 

Hiperspace is like the KV-Cache using by Large Language Models to reference context-windows that are much larger physical memory - like KV-Cache it can take advantage of "Data Processing Units" (DPU) hardware accelerated storage that is close the physical devices  

NVMe Key-Value promises game-changing capacity without increased latency. 

Cache stores add serialization latency for whole graphs that Hiperspace does not need

• Low Latency direct access to information
• Larger Space than virtual memory
• Simpler than a cache service (that need whole objects to be serialized)
• Viewable as a graph without transformation
• History of Elements for point-in-time view of data
• Delta views OLAP aggregates with history
• Horizon (https://www.cepheis.com/hiperspace/horizon) global filtering of context (e.g. approval status)

The runtime is pure open-source GitHub and can be deployed from NuGet (including HiLang compiler for C#)

Roadmap

1. Hiperspace foundation ☑
2. HiLang domain compiler ☑
3. Hiperspace.DB Server (in progress) ☐
4. C++ Hilang target ☐
5. Java Hilang target ☐
6. KV-SSD integration ☐