Data Management Plan Dmp Template

STANDARD OPERATING PROCEDURE: Data Management Plan (DMP) Lifecycle Governance

Document ID: SOP-TR-ENG-409
Effective Date: October 24, 2023
Version: 2.4.0
Review Cadence: Annual
Owner: Julian Vance, Chief Architect, Template Registry

---

1. Executive Summary & Purpose

This Standard Operating Procedure (SOP) defines the institutional requirements, technical workflows, and validation gates for authoring, reviewing, executing, and archiving Data Management Plans (DMPs) within Template Registry. The purpose is to ensure absolute data integrity, metadata compliance, automated security classification, and reproducibility across all internal repositories and client-facing deliverables.

---

2. Scope & Prerequisites

Scope

Applies to all software engineering, data science, research operations, and infrastructure teams operating within Template Registry domains. Covers all structured and unstructured datasets ingested, processed, stored, or decommissioned.

Prerequisites

• Active Enterprise Identity Provider (IdP) credentials with Multi-Factor Authentication (MFA).
• Access to the Template Registry Enterprise DMP Portal (dmp.templateregistry.internal).
• Command-line interface access with git, jq, and aws-cli (v2.x) configured.
• Familiarity with ISO/IEC 27001 data classification standards and FAIR data principles (Findable, Accessible, Interoperable, Reusable).

---

3. Roles & Responsibilities (RACI Matrix)

• Responsible (R): Executes the task and drafts the DMP artifacts.
• Accountable (A): Final sign-off, institutional accountability, and policy enforcement.
• Consulted (C): Subject matter experts providing architectural and compliance input.
• Informed (I): Kept up-to-date on operational status and version updates.

---

4. Step-by-Step Procedure

Phase 1: Ingestion & Metadata Initialization

• 1.1 Access the Template Registry DMP Portal and initialize a new repository instance using the standard template schema (TR-DMP-v2.4).
• 1.2 Assign the definitive Data Owner and Data Steward within the metadata header.
• 1.3 Define the primary data category (e.g., PII, Intellectual Property, Public Telemetry, Confidential Financials) in schema.json.
• 1.4 Establish the cryptographic ledger linkage by registering the dataset's root UUID in the internal cataloging service.

Phase 2: Storage Architecture & Security Classification

• 2.1 Specify storage tiers (Hot, Warm, Cold, Glacier) based on access frequency and performance SLAs.
• 2.2 Configure encryption parameters: Ensure AES-256 is enforced at rest and TLS 1.3 is enforced in transit.
• 2.3 Map access control lists (ACLs) using the Principle of Least Privilege (PoLP); restrict write access strictly to verified service principals.
• 2.4 Execute automated vulnerability and classification scans using the enterprise security scanner:

  tr-sec-scanner --scan-target ./dmp-workspace --enforce-strict
  

Phase 3: FAIR Principles & Interoperability Configuration

• 2.3 Populate the persistent identifier (PID) field using designated DOI or internal URN structures.
• 2.4 Attach semantic definitions, controlled vocabularies, and data dictionaries to all tabular and relational schemas.
• 2.5 Verify that open, machine-readable formats (e.g., Parquet, JSON-LD, CSV) are prioritized over proprietary formats.

Phase 4: Lifecycle Execution & Archival

• 4.1 Define retention schedules in alignment with regulatory mandates (e.g., GDPR, CCPA, corporate governance charters).
• 4.2 Program automated lifecycle transition rules in object storage buckets (e.g., transition to Cold storage after 90 days of inactivity).
• 4.3 Execute final cryptographic shredding validation tests prior to decommissioning datasets.

---

5. Quality Assurance & Pro-Tips

Best Practices

• Shift-Left Compliance: Integrate DMP schema validation into your CI/CD pipeline via GitHub Actions to block unapproved data structures before staging deployment.
• Immutable Versioning: Never overwrite existing DMP iterations; increment semantic versions (MAJOR.MINOR.PATCH) for every structural adjustment.

Common Pitfalls

• Stale Metadata: Failing to update data dictionaries when schema migrations occur will instantly invalidate FAIR compliance metrics.
• Over-Privileging: Granting broad directory-level permissions rather than granular object-level or role-based access controls.

Metric Thresholds

• Metadata Completeness Score: Must evaluate to $\ge 98%$ via the automated validation script before production deployment.
• RPO/RTO Targets: Recovery Point Objective (RPO) $\le 1$ hour; Recovery Time Objective (RTO) $\le 4$ hours for tier-1 operational data.

---

6. Frequently Asked Questions

Q1: What should I do if a dataset's security classification changes mid-project?

A: Immediately update the classification tag in the schema.json configuration file, trigger a manual compliance scan using tr-sec-scanner, and request an expedited review from the Security & Compliance Officer via the DMP Portal.

Q2: How are legacy datasets handled if they lack machine-readable data dictionaries?

A: Legacy datasets must be retrofitted with a minimal-viable schema template within 30 days of discovery, or placed into immediate quarantine storage until a Data Steward completes remediation.