Matthew Morris

  • Nov 19, 2025

How Data 360 Shapes Salesforce Interoperability: Live, Cached, and File Based Patterns Explained

Interoperability is now the centre of gravity in Salesforce architecture. Modern estates span multiple clouds, data platforms, and external systems. The question is no longer whether to integrate. It is how to design the fabric that keeps data consistent, useful, and trustworthy across the enterprise.

At the heart of that fabric are three interoperability patterns. Live federation. Cached federation. File based federation. Each solves a different problem. Each comes with trade offs in latency, cost, freshness, and compute placement. Architects who understand these constraints design platforms that are both modern and manageable.

This also ties directly to a question readers asked after my article on the end of the SObject era. Will SObjects still be needed, and how do they fit with Data 360 DMOs. The answer is yes. SObjects remain the core operational objects inside each org. But they no longer define the enterprise wide model. DMOs sit above them as the unified meaning layer, and your interoperability pattern determines how the two align.

Live Federation

Live federation retrieves data at the moment it is needed, without storing a persistent copy in Salesforce. DMOs define the semantic shape of the data. The live call retrieves the values from a remote system or from Data 360 in real time.

SObjects still anchor the runtime of each app, but live federation avoids unnecessary duplication. Choose this when the business needs fresh values every time: inventory, pricing, eligibility, KYC.

Rule of thumb: Use live federation only when the value of real time accuracy outweighs the risks of latency and dependency on upstream services.

Cached Federation

Cached federation keeps a local copy of external data inside Salesforce or in an adjacent cache. This is the pattern where DMOs and SObjects work hand in hand.

DMOs define the unified view. The cache materialises that view where the compute runs. Sometimes the cache lands in Data 360. Sometimes it materialises into SObjects to support app logic. Think of DMOs as the blueprint and SObjects as the operational footprint.

Use this pattern for product catalogs, account hierarchies, entitlements, or loyalty profiles.

Rule of thumb: Cache when performance and consistency matter more than instant freshness.

File Based Federation

File based federation moves data in bulk, usually through cloud storage layers and batch pipelines. It is the right answer for historical data, planning cycles, reporting, and machine learning workloads.

DMOs define the shape and semantics of the file. SObjects usually are not involved here. The data lands in Data 360 or in cloud storage for downstream compute.

Rule of thumb: Use files for scale and cost efficiency, not operational immediacy.

Where Data 360 Fits When a Customer Already Has a Data Lake

Enterprise architects often challenge Data 360 with a familiar line:
“We already have a data lake. Why do we need this?”

The answer is placement and purpose.

Hyperscaler data lakes are designed for universal storage, analytics, and machine learning compute. Data 360 is designed to deliver a Salesforce ready semantic layer that feeds operational experiences, AI agents, trust frameworks, and real time personalisation. The lake remains the system of record for raw and enriched data. Data 360 becomes the system of meaning for customer facing processes.

  • Use the data lake to store everything, process at massive scale, train models, and run heavy compute.

  • Use Data 360 to unify, govern, align semantics, and serve consistent entities into CRM, service, marketing, activation, and agentic workflows.

The two do not replace each other. They complement. Interoperability patterns are the glue.

Putting It Together

Ask one question. Where should the compute live.

Source system means live federation.
Salesforce runtime means cached federation.
Non immediate workloads mean file based federation.

SObjects run the application.
DMOs run the enterprise.
Data 360 connects the estate.
Interoperability keeps the system coherent.

Further Reading

Salesforce doesn't live alone in landscape of any organisation. Here are some background reading links chosen to help Salesforce customers review similar patterns supported by hyperscaler data providers with Salesforce Data 360.

On data federation and virtualisation

On data lakes and semantic layers