1. Latest Release Home
  2. Guides
  3. Concepts
    1. Aggregate filter
    2. Alias
    3. Apache Avro™
    4. Apache Kafka™
    5. Applied filter
    6. Base table
    7. Bookmark
    8. Business schema
    9. Business schema view
    10. Business schema view column
    11. Business schema view formula column
    12. Column
    13. Connector
    14. Dashboard filter
    15. Dashboard folder
    16. Dashboard runtime filter
    17. Data agent
    18. Data alert
    19. Data destination
    20. Data purge
    21. Data retention
    22. Date system variable
    23. Direct query
    24. Distinct filter
    25. Dynamic field
    26. Expression
    27. External data source
    28. External session variable
    29. External table
    30. Filter expression
    31. Filter expression session variable
    32. Filter option
    33. Formula column
    34. Formula expression
    35. Full load
    36. Global variable
    37. Incorta Analyzer Table
    38. Incorta Analyzer Table column
    39. Incorta Analyzer Table formula column
    40. Incorta Analyzer View
    41. Incorta Analyzer View formula column
    42. Incorta Spark SQL View
    43. Incorta SQL Table
    44. Individual filter
    45. Insight filter
    46. Internal query expression
    47. Internal session variable
    48. Join
    49. Join condition
    50. Join condition column
    51. Join filter
    52. Join filter column
    53. Level-based measure
    54. Load filter
    55. Local data file
    56. Local data folder
    57. Log-based incremental load
    58. Materialized View
    59. Materialized View column
    60. Materialized View formula column
    61. Measure filter
    62. Miscellaneous system variable
    63. Multi-source table
    64. Physical schema
    65. Physical schema table
    66. Physical schema table column
    67. Physical schema table formula column
    68. Physical schema table runtime security filter
    69. PostgreSQL (Incorta SQL) View
    70. Presentation variable
    71. Prompt
    72. Query system variable
    73. Remote table
    74. Runtime business view
    75. Runtime business view formula column
    76. Runtime security filter
    77. Schema evolution
    78. Service Status Observability
    79. Session variable
    80. Shared Storage
    81. Upgrade Readiness
    82. Variable
    83. Visualization
  4. Tools
  5. References
  6. Integrations
  7. Data applications
  8. Data Governance
  9. Release Notes
  10. Incorta Premium
  11. Release Support Policy
  12. Release Model

Concepts → Remote Table

Overview

A Remote table is a physical schema table definition in Incorta that references data stored in an external data lake, allowing access to that data without extracting it into Incorta’s shared storage or loading it into Engine memory. This architecture allows querying large datasets directly from their source locations, such as Google Cloud Storage and Amazon S3.

Remote tables in Incorta are primarily accessed through Materialized Views (MVs) and Spark SQL Views (supported starting 2026.3.0). The main use case for remote tables is to reference large external datasets and perform transformations or aggregations within Incorta, producing smaller, materialized datasets optimized for analytics.

Unlike standard physical tables, remote tables are not extracted or loaded by the Loader Service. Consequently, they are not directly available to the Analytics Service for querying. Instead, they must be referenced through MVs or Spark SQL Views, where the data is processed and materialized before being consumed by dashboards or external BI tools.

This approach helps reduce storage and memory usage as only the final transformed or aggregated results are loaded into Incorta Engine, rather than the full raw dataset.

Key benefits and use cases

  • Aggregating large data without full extraction: Remote tables enable referencing very large datasets and creating aggregated materialized views or Spark SQL views without loading the full raw data into Incorta.
  • Reducing storage duplication: Duplicating large datasets in shared storage can consume significant disk space and memory. Remote tables eliminate unnecessary duplication when only a subset or an aggregated dataset is needed.
  • Data lake–first architectures: Organizations maintaining centralized data lakes can share datasets with Incorta without disrupting existing storage architecture.
  • External BI tool integration: External BI tools can access remote data via Incorta’s SQL interfaces without physically loading it into Incorta.

Supported connectors

Incorta supports remote tables for the following data lake connectors:

How remote tables work

  1. Create a data lake data source with appropriate cloud storage credentials.

  2. Define a physical schema table in Incorta that references the data source, turn on the Remote toggle, and specify the remote data file type.

    Notes

    Before 2025.7, remote tables support text-based, ORC, and Parquet files only, while they start supporting Delta tables in 2025.7. They support union files starting with 2026.3.0.

  3. Do one of the following to access the remote data:

    • Create an MV that references the remote table, and load the MV to enable Analytics Service queries.
    • Create a Spark SQL view (available starting 2026.3.0) for quick insights without materializing data.
    • Use external BI tools, such as Power BI and Tableau, to query remote data directly via the standard SQL Interface or Advanced SQL Interface (supported starting 2026.3.0).

Remote tables vs. performance-optimized physical tables

AspectRemote tablesPerformance-optimized physical tables
Data LocationExternal storageIncorta memory
LoadingNo extractionFull extraction and loading
Analytics Service AccessVia MVs or Spark SQL views onlyDirect access
Memory UsageMinimalHigh (data in memory)
Query PerformanceDepends on the external system and SparkOptimized in-memory queries
Storage RequirementsNone in IncortaFull dataset stored
Use CaseLarge datasets, selective accessFrequently queried data, fast analytics

Known limitations

  • The Analytics Service cannot query remote tables directly. To use remote data in dashboards, insights, or business views, you must first create a materialized view or Spark SQL view that references the remote table
  • Standard incremental load capabilities don't apply to remote tables.
  • The performance of queries against remote tables (via MVs or Spark SQL views) depends on network latency and source system availability and performance.
  • Remote tables do not benefit from Incorta's in-memory processing and DDM processes unless materialized as MVs.