Historical Access

The ReadRaw service allows a Client to request unprocessed, timestamped values from a historian for a specified time domain. The Server returns a sequence of HistoryData values in chronological order, preserving the original sampling fidelity.

• Bounded requests: Clients specify startTime and endTime to limit the query scope.
• Return bounds: The Server can return bounding values outside the requested interval to provide context.
• Max return values: Clients can cap the total number of returned samples to prevent memory exhaustion.
• Continuation points: For large datasets, the Server returns a continuation point enabling paginated retrieval.

The ReadProcessed service computes statistical aggregates over raw historian data, offloading computation from the Client to the Server. This dramatically reduces network payload when only summary insights are required.

• Standard aggregates: Includes Interpolative Bounding Values, Average, TimeAverage, Minimum, Maximum, Range, Count, and Standard Deviation.
• Resample interval: Clients define a processing interval over which each aggregate is calculated.
• Stepped interpolation: The Server treats data as a stepped function between samples for accurate time-weighted calculations.
• Annotation support: Annotations indicating data quality can be included alongside computed aggregates.

The ReadModified service retrieves historical values along with metadata indicating when and by whom a value was manually overwritten. This is critical for regulated industries requiring complete audit trails.

• ModificationInfo: Each returned value includes ModificationTime and UserName of the operator who made the change.
• Original value preservation: The Server retains the original value before modification for compliance.
• Sequential history: Both original and modified values are returned in chronological order.
• Audit integrity: Supports 21 CFR Part 11 and similar regulatory requirements for electronic records.

The ReadEvent service allows Clients to query historical event logs stored by the Server, filtering by event type, severity, time range, and source Node. This enables forensic analysis of past alarms and system conditions.

• Event filter: Clients submit an EventFilter defining the selection criteria and the event fields to return.
• Time domain: Queries are bounded by startTime and endTime.
• Event types: Supports AlarmCondition, SystemEvent, AuditEvent, and custom event types.
• Ordered delivery: Events are returned in chronological order based on their timestamp.

The UpdateAnnotations service enables Clients to attach human-readable comments to specific historical data points. Annotations provide operational context that raw timestamps and values cannot convey.

• Annotation message: A free-text string explaining the significance of a data point.
• User identification: The Server records the UserName associated with the annotation.
• Timestamp association: Annotations are linked to a specific value timestamp.
• Retrieval integration: Annotations are returned alongside raw data when requested via ReadRaw with annotation flags enabled.

For queries that return large datasets exceeding a single response limit, the Server issues an opaque ContinuationPoint. The Client passes this token in a subsequent request to retrieve the next batch of results.

• Session affinity: Continuation points are valid only within the Session that created them.
• Timeout management: Unused continuation points are released after a Server-defined timeout.
• Release service: Clients should call ReleaseContinuationPoints to free Server resources when pagination is abandoned.
• Stateless pagination: Enables efficient retrieval of multi-gigabyte historian archives without overwhelming Client memory.