## Performance evidence update / 18 September 2026 MODALITY PERFORMANCE / EVIDENCE REVIEWED 18 SEPTEMBER 2026 Follow five relationships in one query. Historical service comparison September 8, 2026 campaign queries / second Higher is better · linear scale from zero 8DB 23,300 Eight workers · median Neo4j 4,134 32 connections · best reported round M1 laptop; one million persons and five million directed relationships. Simplified five-friend query, different clients and returned-data contracts. Original traversal log is unavailable; retained report records successful checks. Resident reads, not durable ingestion or a full LDBC score. Read the results, comparison and methods → What happens when models work together? See the separate graphs + time series + spatial benchmark against SurrealDB → 01 Exact affected set 02 Depth, fanout & p95 03 Updates & source assembly Include memory, storage, energy and hardware writes [Read the results, comparison and methods →](/journal/8db-graph-lookups-ldbc) [See the separate graphs + time series + spatial benchmark against SurrealDB →](/modalities#multimodal-performance) ## Interactive evaluation example / 18 September 2026 [Explore the visual example](/outcomes/change-impact#example). [Download the semantic fixture and acceptance checklist](/evaluation/change-impact.json). Browser illustration only; not a live 8DB runtime or benchmark. Pin the engine build, platform, adapters and protection scope for an assisted evaluation. ## Implementation update — 18 September 2026 Implementation snapshot / Reviewed 18 September 2026 Follow changed evidence to the work that needs review. Source-reviewed research fixture The research evidence twin has a concrete dependency-withdrawal example. Its query source follows recorded derivation edges; the regression expects two dependent statements and excludes independently supported work. This provides a specific application of relationships and evidence rules. [Inspect the dependency example →](/outcomes/change-impact#example) API, release scope and evidence limits Archived Rust interfaces: retraction_cascade, direct_premises and confidence_ceiling_path. Source snapshot dated 12 September 2026; reviewed 18 September. Source review is not a new test run or a generally released change-management service. Customer rules, current build, APIs and deployment need matching qualification. The fixture follows recorded support; it does not prove causation or assess all possible independent evidence. --- # Graphs & hierarchies | 8DB Product reference. Reviewed 18 September 2026. Implementation and evaluation scope are stated below. Explore the models (https://8braid.com/modalities) / 8DB Graphs & hierarchies Follow the relationship to a better answer. Represent connected entities, traverse typed relationships and bring the relevant paths back into the application. Explore the capabilities See a complete workload Connect · traverse · follow Model / Graphs Conceptual diagram [Diagram: graphs representation: conceptual diagram] Entities form a network. The highlighted path follows a relationship through connected records. Capabilities What changes Input to value Alternatives Performance Adoption Related articles Reference Capabilities / documented scope Ask how things connect. Then follow the consequence. 01 / Relationships Give a connection meaning Typed, directed relationships distinguish an asset that contains a component from a document that describes it. Your application defines the meaning of each relation. 02 / Traversal Follow the question either way Supported bidirectional traversal lets a workflow move from an observation to its asset, or from an asset toward related observations. 03 / Structure Represent dependencies and hierarchies Use relationships to represent parentage and recorded dependencies. The application determines which relations express support, ownership or an operational dependency. Documentary review, 18 September 2026 UTC. Qualify traversal depth, fanout, persistence and consistency on the selected build. Graph representation does not establish general Cypher compatibility or automatic dependency maintenance. Why consider 8DB / connected native models Every connection has a bigger story. Follow a relationship into the material that helps a person assess it. [Diagram: documents representation: conceptual diagram] documents Source passages Inspect what supports the relationship. Explore model ↗ (https://8braid.com/modalities/documents) [Diagram: time series representation: conceptual diagram] time-series Dated observations Understand the state around the change. Explore model ↗ (https://8braid.com/modalities/time-series) [Diagram: hypergraphs representation: conceptual diagram] hypergraphs Richer relationships Represent several participants in one event. Explore model ↗ (https://8braid.com/modalities/hypergraphs) Conceptual model connections. 8DB offers these native representations; fewer handoffs and simpler application assembly are opportunities to test on the supported combination. Input → usable value / interactive design example See a change travel through the evidence. 0 1 A finding changes → 0 2 Follow the dependencies → 0 3 Check the observations → 0 4 Explain the impact → [Diagram: documents representation: conceptual diagram] Step 1 / documents Inspection revised A source finding changes for Pump P-17. Keep the source passage and the affected asset identity together. Next: Follow the dependencies Illustrative application design, not an executed benchmark. Qualify the composed path, including external models, authorization, persistence and update visibility. What belongs in the complete-workflow comparison? Measure from the declared input state to the usable output, including required parsing, embeddings, storage, index readiness, querying and result assembly. Report database-stage timing separately. Include memory, storage, network and measured energy when relevant. Market context Compare the relationship and the work around it. Approach What it brings What to compare with 8DB Neo4j (https://neo4j.com/docs/getting-started/appendix/graphdb-concepts/) A property-graph model with typed relationships, properties, indexes and constraints. Required graph operators, query tooling, traversal behavior and how the application uses source evidence. SurrealDB (https://surrealdb.com/docs/learn/data-models) Graph relationships within a broader data-model offering. Equivalent connected workflows, supported representations and operating requirements. 8DB Typed relationships alongside documents, observations and additional native structures. The complete route from incoming evidence to a correct, inspectable impact report. Property graphs can already represent rich context. 8DB's case should rest on the useful combination of operations and deployment characteristics your workload can demonstrate. Vendor documentation reviewed 18 September 2026 UTC. This is a selection guide, not a traversal-performance ranking. Performance & resources Measure the work that delivers the answer. Correct reachable result set; p95 at declared depth/fanout; ingestion and index time; update visibility; source reconstruction; memory/disk; cold and warm behaviour. Published results apply to the named operation, build and workload. The evaluation below separates a component result from the database service and the complete application outcome. 01 Traversal correctness Expected paths and returned records at declared depth and fanout. Include cycles, duplicates and changing relationships. 02 Useful response Time from a revised finding becoming queryable to a complete, inspectable impact report. 03 Operating cost Load and update time, peak memory, stored bytes, physical writes and measured energy per successful job. Explore the published benchmark evidence (https://8braid.com/benchmarks) What belongs inside the timer and resource budget? For input-to-value testing, declare the input state and the consumer-ready output first. Include parsing, derivation, storage/protection, index readiness, querying, result assembly and delivery as applicable. Identify amortized setup and any work deferred past the finish line. Report database-stage and full-job results separately. Do not add unrelated medians or p95 values, or multiply component speedups into a claimed workflow result. Energy needs measured joules; reduced handoffs alone do not establish a saving. Choose the path that fits your deployment. Verify the supported API/SDK, data format, persistence and indexing behaviour for the release you plan to use. Match the operation-by-modality protection profile and device qualification to the workload. Post-quantum protection (https://8braid.com/pqc) · Plan a workload evaluation (https://8braid.com/build) Related articles Take the next step into the thinking. Mechanism & application When Evidence Changes, Know Which Decisions Need Review (https://8braid.com/journal/a-proof-is-a-graph-where-confidence-only-flows-downhill) Explore the ideas and implementation context behind this capability. Mechanism & application Show Your Work: Provenance as a Query, Not a Forensic Reconstruction (https://8braid.com/journal/show-your-work-provenance-as-a-query) Explore the ideas and implementation context behind this capability. Reference / Reviewed 18 September 2026 UTC Readable, shareable and traceable. This reference explains 8DB’s public modality description (https://8braid.com/modalities#explore) with an explicit workload and selection criteria. It is a documentary review, not fresh execution of every operator or platform. Public sources include the linked vendor documentation in the comparison above. Additional context: 8DB modality scope (https://8braid.com/modalities) ; Neo4j graph concepts (https://neo4j.com/docs/getting-started/appendix/graphdb-concepts/) ; the related articles above. Internal source leads and outstanding release checks remain in the editorial record. Download this reference (https://8braid.com/modality-reference/graphs.md) Evaluate the question your application needs to answer. Bring the data shapes, input state, expected output and the systems you use today. Define a comparison around that complete job. Discuss your workload (https://8braid.com/build)