Circular loop data model (CLDM): a novel approach to simplifying data management for rapid prototyping

Abstract

We introduce the Circular Loop Data Model (CLDM), a lightweight data model and domain-specific language (DSL) for embedded, offline-first applications and rapid prototyping. CLDM organises application state as named loops of parallel string-valued segments, persisted in a single BSON file with file-level locking for cross-process concurrency. The DSL exposes six core commands ( FORGE , SEGMENT , CRAFT , EXTRACT , COMBINE ) that map directly to common application workflows without requiring SQL or upfront schema definition. Here, lightweight refers to deployment and operational overhead: no server process, no schema migration, and single-file distribution. We evaluate CLDM across write, read, aggregation, file-size, and concurrency benchmarks on a commodity laptop. Key results: in-process embedding delivers a mean of 2.3 ms per CRAFT operation; CLI invocation costs approximately 214 ms per command due to Python interpreter startup; and CLDM reads 500,000 records in 0.183 s, competitive with manual BSON file parsing. The primary limitation is O(N 2 ) write complexity under the save-on-write policy: inserting 10,000 records sequentially requires 444 s, approximately \(38{,}000\times\) slower than SQLite with transactions. CLDM is therefore suited to low-volume local workloads and prototyping scenarios, not bulk-insert or high-frequency logging.