/datasphere-space-template

Generate a Datasphere space configuration template based on the requested type.

Available Space Types

development

Development space for modelers with full build permissions and moderate resources.

analytics

Production analytics space with optimized settings for consumption and reporting.

integration

Integration space focused on data loading, replication, and ETL processes.

sandbox

Sandbox space with minimal resources for experimentation and learning.

---

Template: development

Space Configuration (Development)

{
  "space": {
    "spaceId": "DEV_<PROJECT>",
    "businessName": "Development - <Project Name>",
    "description": "Development environment for data modeling and testing",
    "settings": {
      "storage": {
        "diskSize": "50 GB",
        "recommendation": "Start with 50GB, expand as needed based on data volume"
      },
      "memory": {
        "inMemory": "16 GB",
        "recommendation": "Sufficient for development workloads"
      },
      "priority": 4,
      "priorityNote": "Medium priority (1=highest, 8=lowest)"
    }
  }
}

Recommended Roles

Role	Users	Purpose
DW Space Administrator	1-2	Space configuration and management
DW Modeler	3-5	Create views, models, flows
DW Integrator	1-2	Configure connections, run flows
DW Viewer	As needed	View shared objects

Best Practices

1. Naming Convention: Use DEV_ prefix for all development spaces
2. Connections: Create development-specific connections (don't share with production)
3. Data Access Controls: Implement early to test security
4. Content Transport: Export regularly for backup
5. Documentation: Maintain README in space for onboarding

Space Properties to Configure

Space Properties:
  Allow Public Objects: true        # Enable sharing to other spaces
  Enable Audit Logging: true        # Track changes for debugging
  Data Integration Monitor: Enabled # Monitor flows and tasks
  Database User Access: Optional    # Create if external tools needed

---

Template: analytics

Space Configuration (Analytics/Production)

{
  "space": {
    "spaceId": "PROD_ANALYTICS",
    "businessName": "Analytics - Production",
    "description": "Production analytics space for SAC and external consumption",
    "settings": {
      "storage": {
        "diskSize": "200 GB",
        "recommendation": "Size based on analytic model data volumes"
      },
      "memory": {
        "inMemory": "64 GB",
        "recommendation": "Higher for concurrent query performance"
      },
      "priority": 2,
      "priorityNote": "High priority for production analytics"
    }
  }
}

Recommended Roles

Role	Users	Purpose
DW Space Administrator	1-2	Space management only
DW Modeler	2-3	Limited for maintenance
DW Viewer	Many	Consume analytics

Best Practices

1. Naming Convention: Use PROD_ prefix for production spaces
2. View Persistence: Enable for frequently accessed views
3. Statistics: Keep table statistics up-to-date
4. Monitoring: Set up alerts for resource usage
5. Access Controls: Production data access controls required

Analytic Model Optimization

Optimization Settings:
  View Persistence:
    - Snapshot: For slowly changing data (refresh daily/weekly)
    - Real-time: For frequently changing data (use sparingly)

  Partition Strategy:
    - Time-based: For historical analysis (Year/Month)
    - Range: For large datasets with clear boundaries

  Statistics:
    - Enable automatic statistics collection
    - Schedule during off-peak hours

---

Template: integration

Space Configuration (Integration)

{
  "space": {
    "spaceId": "INT_<SOURCE>",
    "businessName": "Integration - <Source System>",
    "description": "Data integration hub for <source> data ingestion",
    "settings": {
      "storage": {
        "diskSize": "500 GB",
        "recommendation": "Large storage for raw/staging data"
      },
      "memory": {
        "inMemory": "32 GB",
        "recommendation": "Moderate memory for transformation"
      },
      "priority": 3,
      "priorityNote": "Medium-high for ETL workloads"
    }
  }
}

Recommended Roles

Role	Users	Purpose
DW Space Administrator	1-2	Space management
DW Integrator	3-5	Primary role for integration work
DW Modeler	2-3	Staging layer modeling
DW Viewer	Limited	Troubleshooting only

Best Practices

1. Naming Convention: Use INT_ prefix with source system name
2. Layered Architecture:
   • raw_ prefix: Unchanged source data
   • stg_ prefix: Cleansed staging data
   • out_ prefix: Output to analytics spaces
3. Task Chains: Orchestrate complex ETL workflows
4. Error Handling: Configure email notifications
5. Logging: Enable detailed flow logging

Integration Architecture

Data Layers:
  Raw Layer:
    - Tables: raw_<source>_<entity>
    - Purpose: Land data unchanged
    - Retention: Keep for audit trail

  Staging Layer:
    - Views: stg_<source>_<entity>
    - Purpose: Cleanse, validate, transform
    - Pattern: Graphical views with filters

  Output Layer:
    - Views: out_<entity>
    - Purpose: Share to analytics spaces
    - Semantic: Set to Relational Dataset

---

Template: sandbox

Space Configuration (Sandbox)

{
  "space": {
    "spaceId": "SANDBOX_<USER>",
    "businessName": "Sandbox - <User/Team>",
    "description": "Experimental sandbox for prototyping",
    "settings": {
      "storage": {
        "diskSize": "10 GB",
        "recommendation": "Minimal for experimentation"
      },
      "memory": {
        "inMemory": "8 GB",
        "recommendation": "Basic allocation"
      },
      "priority": 8,
      "priorityNote": "Lowest priority (won't impact production)"
    }
  }
}

Recommended Roles

Role	Users	Purpose
DW Space Administrator	1	Space owner
DW Modeler	1-2	Experiment with modeling
DW Integrator	1	Test integrations

Best Practices

1. Naming Convention: Use SANDBOX_ prefix with owner identifier
2. Isolation: No connections to production systems
3. Cleanup: Regular cleanup policy (monthly)
4. Sample Data: Use sample datasets or anonymized data
5. Documentation: Track experiments and learnings

Sandbox Policies

Sandbox Policies:
  Data:
    - Use sample content package
    - Anonymize production data if needed
    - No real PII/sensitive data

  Resources:
    - Monitor for runaway processes
    - Auto-kill long-running queries (30 min)

  Lifecycle:
    - Review monthly
    - Archive unused sandboxes quarterly

---

CLI Commands for Space Management

Create Space

datasphere spaces create \
  --space-id "DEV_PROJECT" \
  --business-name "Development - Project" \
  --disk-storage 50 \
  --memory 16

Allocate Storage

datasphere spaces update \
  --space "DEV_PROJECT" \
  --disk-storage 100

Assign Users

datasphere spaces users add \
  --space "DEV_PROJECT" \
  --user "user@company.com" \
  --role "DW Modeler"

---

Provide the appropriate template based on the user's requested type. Customize the space ID, names, and sizes based on their specific project requirements.