Polaris is the distributed SQL engine behind Fabric Data Warehouse, bringing SQL Server query processing to distributed compute over shared OneLake storage.
Based on the Polaris papers, this session explains how T-SQL in SQL Pools is compiled by the Front End nodes, optimized into distributed physical plans, orchestrated by the Distributed Computing Platform as workflow DAGs, and executed by Back End compute nodes over data cells. We connect that engine design to Parquet storage, Delta Lake log publishing, query decorrelation, and snapshot isolation, and show what it means for data movement, concurrency, plan quality, and performance at scale.

Microsoft Research and engineering teams continue to develop Polaris. Synapse Serverless SQL pool was its initial read-only implementation for querying external data. Fabric Warehouse is its first full read/write implementation, with support for transactions across statements and tables.

The scheduled session is 2x90 minutes (level 500). The deck includes worked examples, diagram walkthroughs and discussion.

Slides

The slides are available through the Download slides button.

Benchmarks and query plans

Explore the recorded TPC-H, TPC-DS and DSB timings from the SF100 Fabric Warehouses on F64. All 177 query plans include the SQL, operator details and source XML. The six presentation examples are marked Demo.

Benchmark results · Interactive query plans

Research papers

The reference numbers and IEEE citation style match the slides. Each title links to the paper, with a DOI where available. Expand Full author list to see all authors in their published order.

Start with the Polaris architecture paper , then follow the topics you want to explore further.

The author index and research connections links all 148 authors to their papers. Its paper-pair matrix shows the shared authors across Gray Systems Lab, the SQL/Fabric teams, Microsoft Research and university collaborators. It also gives the background on Jim Gray and the lab named after him.

[1] J. Aguilar-Saborit et al., “Polaris: The Distributed SQL Engine in Azure Synapse,” Proc. VLDB Endow., vol. 13, no. 12, pp. 3204–3216, 2020. doi: 10.14778/3415478.3415545.

Full author list (49)

Josep Aguilar-Saborit, Raghu Ramakrishnan, Krish Srinivasan, Kevin Bocksrocker, Ioannis Alagiannis, Mahadevan Sankara, Moe Shafiei, Jose Blakeley, Girish Dasarathy, Sumeet Dash, Lazar Davidovic, Maja Damjanic, Slobodan Djunic, Nemanja Djurkic, Charles Feddersen, Cesar Galindo-Legaria, Alan Halverson, Milana Kovacevic, Nikola Kicovic, Goran Lukic, Djordje Maksimovic, Ana Manic, Nikola Markovic, Bosko Mihic, Ugljesa Milic, Marko Milojevic, Tapas Nayak, Milan Potocnik, Milos Radic, Bozidar Radivojevic, Srikumar Rangarajan, Milan Ruzic, Milan Simic, Marko Sosic, Igor Stanko, Maja Stikic, Sasa Stanojkov, Vukasin Stefanovic, Milos Sukovic, Aleksandar Tomic, Dragan Tomic, Steve Toscano, Djordje Trifunovic, Veljko Vasic, Tomer Verona, Aleksandar Vujic, Nikola Vujic, Marko Vukovic, and Marko Zivanovic.

[2] J. Aguilar-Saborit et al., “Extending Polaris to Support Transactions,” 2024, arXiv:2401.11162. doi: 10.48550/arXiv.2401.11162. Author preprint.

Full author list (18)

Josep Aguilar-Saborit, Raghu Ramakrishnan, Kevin Bocksrocker, Alan Halverson, Konstantin Kosinsky, Ryan O’Connor, Nadejda Poliakova, Moe Shafiei, Taewoo Kim, Phil Kon-Kim, Haris Mahmud-Ansari, Blazej Matuszyk, Matt Miles, Sumin Mohanan, Cristian Petculescu, Ishan Rahesh-Madan, Emma Rose-Wirshing, and Elias Yousefi.

[3] R. Alotaibi et al., “Towards Query Optimizer as a Service (QOaaS) in a Unified LakeHouse Ecosystem: Can One QO Rule Them All?,” in Proc. CIDR, 2025. [Online]. Available: Proceedings PDF. Author preprint. Original Figures 1 and 3 show the proposed shared optimizer architecture and measured Spark prototype. Substrait supplies the plan exchange format; engine capabilities, costs and workload feedback guide the proposed service.

Full author list (16)

Rana Alotaibi, Yuanyuan Tian, Stefan Grafberger, Jesús Camacho-Rodríguez, Nicolas Bruno, Brian Kroth, Sergiy Matusevych, Ashvin Agrawal, Mahesh Behera, Ashit Gosalia, Cesar Galindo-Legaria, Milind Joshi, Milan Potocnik, Beysim Sezgin, Xiaoyu Li, and Carlo Curino.

[4] H. Zhao et al., “I Can’t Believe It’s Not Yannakakis: Pragmatic Bitmap Filters in Microsoft SQL Server,” in Proc. CIDR, 2026. [Online]. Available: Proceedings PDF.

Full author list (10)

Hangdong Zhao, Yuanyuan Tian, Rana Alotaibi, Bailu Ding, Nicolas Bruno, Jesús Camacho-Rodríguez, Vassilis Papadimos, Ernesto Cervantes Juárez, Cesar Galindo-Legaria, and Carlo Curino.

[5] G. Yeo et al., “ZipFlow: a Compiler-based Framework to Unleash Compressed Data Movement for Modern GPUs,” 2026, arXiv:2602.08190. doi: 10.48550/arXiv.2602.08190. Author preprint.

Full author list (8)

Gwangoo Yeo, Zhiyang Shen, Wei Cui, Matteo Interlandi, Rathijit Sen, Bailu Ding, Qi Chen, and Minsoo Rhu.

[6] N. Bruno et al., “Unified Query Optimization in the Fabric Data Warehouse,” in Proc. ACM SIGMOD Companion, 2024, pp. 18–30. doi: 10.1145/3626246.3653369.

Full author list (9)

Nicolas Bruno, César Galindo-Legaria, Milind Joshi, Esteban Calvo Vargas, Kabita Mahapatra, Sharon Ravindran, Guoheng Chen, Ernesto Cervantes Juárez, and Beysim Sezgin.

[7] N. Bruno, C. Galindo-Legaria, and M. Joshi, “Query Decorrelation in the Fabric Data Warehouse,” in Proc. ACM SIGMOD Companion, 2025, pp. 297–309. doi: 10.1145/3722212.3724448.

Full author list (3)

Nicolas Bruno, César Galindo-Legaria, and Milind Joshi.

[8] N. Bruno, E. Cervantes Juárez, C. Galindo-Legaria, M. Joshi, J. A. López Magaña, and J. Aguilar Saborit, “Bitmap Filtering in the Fabric Data Warehouse,” in Proc. ACM SIGMOD Companion, 2026, pp. 227–238. doi: 10.1145/3788853.3803075.

Full author list (6)

Nicolas Bruno, Ernesto Cervantes Juárez, César Galindo-Legaria, Milind Joshi, José Alberto López Magaña, and Josep Aguilar Saborit.

[9] M. Interlandi et al., “CoddSpeed: Hardware Accelerated Query Processing in Microsoft Fabric,” in Proc. ACM SIGMOD Companion, 2026, pp. 359–372. doi: 10.1145/3788853.3803077.

Full author list (64)

Matteo Interlandi, Nicolas Bruno, Brandon Haynes, Carlo Curino, Rathijit Sen, Yinan Li, Kaushik Rajan, Bailu Ding, Lukas M. Maas, Wei Cui, Kevin Gaffney, Mingsheng Hong, Brian Kroth, Sampath Rajendra, Peng Cheng, Surajit Chaudhuri, Johannes Gehrke, Raghu Ramakrishnan, Lidong Zhou, Momin Al-Ghosien, Craig Peeper, Marius Dumitru, Conor Cunningham, Kevin Bocksrocker, Prasanna Sundar, Ron Boskovic, YongChul Kwon, Marko Zivanovic, Runbin Shi, Krystian Sakowski, Denis Lamtsov, Josep Aguilar Saborit, Krish Srinivasan, Adarsh Kapil, Andrew Putnam, Anudeep Kambapu, Aparna Konduri, Aaron Landy, Aaron Moore, Aaron Pitman, Artem Oks, Ashit Gosalia, Babu Kandimalla, Blake Pelton, Bogdan Crivat, Cesar Galindo-Legaria, Chidamber Kulkarni, Jesus Camacho Rodriguez, Evgeny Babin, Hans Lehnert Marino, Igor Konev, Israel Arroyo, Luis Penaranda, Mandar Datar, Morten Borup Petersen, Nicholas Simmons, Parminder Poonian, Pravija Danda, Rob Rydberg, Esteban Calvo, Ronak Bajaj, Sai Sumanth Kammal Shetty, Sharanya Bhat, and Zach Iracheta.

[10] Z. Huang et al., “GPU Acceleration of SQL Analytics on Compressed Data,” Proc. VLDB Endow., vol. 19, no. 3, pp. 320–333, 2025. doi: 10.14778/3778092.3778095. Presented at VLDB 2026. Author preprint.

Full author list (8)

Zezhou Huang, Krystian Sakowski, Hans Lehnert, Wei Cui, Carlo Curino, Matteo Interlandi, Marius Dumitru, and Rathijit Sen.

[11] B. Ding, V. Narasayya, and S. Chaudhuri, “Extensible Query Optimizers in Practice,” Found. Trends Databases, vol. 14, no. 3–4, pp. 186–402, 2024. doi: 10.1561/1900000077.

Full author list (3)

Bailu Ding, Vivek Narasayya, and Surajit Chaudhuri.

[12] P. G. Selinger, M. M. Astrahan, D. D. Chamberlin, R. A. Lorie, and T. G. Price, “Access Path Selection in a Relational Database Management System,” in Proc. ACM SIGMOD, 1979, pp. 23–34. doi: 10.1145/582095.582099. The System R paper that introduced the plan-selection approach and interesting orders discussed in the session.

Full author list (5)

P. Griffiths Selinger, M.M. Astrahan, D.D. Chamberlin, R.A. Lorie, and T.G. Price.

[13] H. Zhao, R. Sen, C. Peeper, and M. Interlandi, “Rubberband: Memory-Elastic, Skew-Tolerant GPU Hash Joins,” in Proc. ADMS, 2026. [Online]. Available: Proceedings PDF.

Full author list (4)

Hangdong Zhao, Rathijit Sen, Craig Peeper, and Matteo Interlandi.

[14] W. Cui, P. Cheng, C. Curino, R. Sen, and M. Interlandi, “TQP++: Bridging ML Compilers and Analytical Query Processing on GPUs,” Proc. VLDB Endow., vol. 19, no. 12, pp. 4116–4129, 2026. doi: 10.14778/3827998.3828020.

Full author list (5)

Wei Cui, Peng Cheng, Carlo Curino, Rathijit Sen, and Matteo Interlandi.

[15] K. Rajan et al., “Scalable GPU Acceleration of Scalar Functions in Analytical Databases: Compilation, Benchmarking, and Optimization,” Proc. VLDB Endow., vol. 19, no. 7, pp. 1441–1454, 2026. doi: 10.14778/3801059.3801061.

Full author list (11)

Kaushik Rajan, Sampath Rajendra, Momin Al-Ghosien, Nicolas Bruno, Carlo Curino, Matteo Interlandi, Yinan Li, Lukas M. Maas, Craig Peeper, Surajit Chaudhuri, and Johannes Gehrke.

[16] B. Wu, W. Cui, C. Curino, M. Interlandi, and R. Sen, “Terabyte-Scale Analytics in the Blink of an Eye,” Proc. VLDB Endow., vol. 19, no. 2, pp. 141–155, 2025. doi: 10.14778/3773749.3773754.

Full author list (5)

Bowen Wu, Wei Cui, Carlo Curino, Matteo Interlandi, and Rathijit Sen.

Formats and documentation

Apache Parquet file format .
Databricks dynamic file pruning and its join-key filtering design : a comparison with Fabric’s runtime distributed bitmap filters.
Delta Lake transaction log protocol .
Partitioning for Delta tables : directory groups of Parquet files and partition pruning.
Delta Lake deletion vectors and a worked transaction-log example .
Warehouse transactions , Delta-log publication and OneLake Iceberg metadata .
Automatic log checkpointing for Fabric Warehouse : asynchronous summaries of the internal table log after every ten transactions.
Delta UniForm : additional format metadata over shared Parquet data. Fabric exposes its Iceberg representation through OneLake metadata virtualization.
Substrait : relational operators , field references and function declarations .
Substrait community governance and Apache 2.0 license . This independent open-source project has contributors affiliated with Google, IBM, NVIDIA, Intel, Voltron Data and Microsoft. Fabric adopts the community’s plan format.
Substrait output mapping with emit and binary serialization .
Custom SQL pools and workload management in Fabric Data Warehouse.

The worked Substrait plan contains the electricity usage example used in the slides. It describes SELECT account_id FROM EnergyUsage WHERE kwh > 100 with field references, function declarations and the final output mapping. This small example uses whole kWh in signed 64-bit integer fields and the Substrait v0.103.0 schema.

Fabric Data Warehouse behavior

These Microsoft Learn pages cover the product behavior used in the examples and connect it to the mechanisms in the papers.

Synapse SQL features : the initial Serverless SQL implementation offers read-only table access and result export through CREATE EXTERNAL TABLE AS SELECT (CETAS); its transactions apply to metadata objects.
Warehouse and SQL analytics endpoints and warehouse architecture .
Data clustering and V-Order : physical layout and the cost of writing data for repeated reads.
Memory and SSD caching and result set caching : different forms of reuse during query execution.
Statistics and performance guidelines : estimates, plan costs, ingestion and parallel execution.
SHOWPLAN XML : inspect the chosen operators and compilation estimates. Query Insights records measured request-level metrics after execution. The historical comparison uses Synapse Dedicated EXPLAIN .
Estimated Graphical Query Plan announcement : Microsoft announced an interactive estimated plan in the Fabric SQL query editor, with public preview coming soon.
Actual execution plan XML: gated preview supplies observed rows and operator execution metrics. The original Microsoft FabCon monitoring demo captures the XML of an active query in Fabric and opens it in SSMS. The slides retain this animated demo. The benchmark plans captured for this Polaris presentation contain compilation estimates.
Query acceleration and its FAQ : eligible SQL work, preview scope, Power BI DirectQuery and billing.
Warehouse data types and CU usage reporting : text representation, accelerated execution eligibility and consumed capacity.
Delta Lake REORG and table maintenance across Fabric workloads : compaction, removal of soft-deleted rows and cleanup after retention.
SQL analytics endpoint performance : metadata sync, scan parallelism and Lakehouse table maintenance.
Warehouse troubleshooting and query blocking : runtime evidence, table-level conflicts and schema-lock waits.
Time travel and table clones : historical queries and writable table branches over retained data.
Warehouse Snapshots and snapshot management : a read-only reporting endpoint whose timestamp can advance after a transformation run.
Restore points , retention and retention configuration : catalog recovery, historical availability and the storage lifetime of referenced files.
OneLake file events : path-based triggers alongside the Delta publication controls.

The presentation also uses the supplied slides from W29, Data Warehousing for the AI Era: Deep Dive into High-Performance Analytics and GPU-acceleration, presented by Conor Cunningham, Maraki Ketema and Filip Popovic at FabCon Europe on 30 September 2026. These provide the aggregate-pushdown CPU example, production scan telemetry, learned resource estimation and the updated DirectQuery comparisons. The CU-consumption examples on page 33 are illustrative estimates.

Supporting references

The documentation, announcements, session material and artwork below use the same reference numbers as the deck. The Microsoft Learn links describe the Fabric features used in the examples.

Full supporting bibliography [17–108]

[17] Microsoft, “Architecture of Fabric Data Warehouse,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/architecture.

[18] Microsoft, “Configure data retention,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/how-to-configure-retention.

[19] Microsoft, “Create and manage a Warehouse Snapshot,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/create-manage-warehouse-snapshot.

[20] Microsoft, “Create streams from OneLake events,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/real-time-hub/create-streams-fabric-onelake-events.

[21] Microsoft, “Custom SQL pools,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/custom-sql-pools.

[22] Microsoft, “Data clustering in Fabric Data Warehouse (preview),” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/data-clustering.

[23] Microsoft, “Data retention in Fabric Data Warehouse,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/data-retention.

[24] Microsoft, “Data types in Fabric Data Warehouse,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/data-types.

[25] Microsoft, “DBCC SHOW_STATISTICS (Transact-SQL),” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/sql/t-sql/database-console-commands/dbcc-show-statistics-transact-sql?view=fabric.

[26] Microsoft, “Delta Lake logs in Warehouse,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/query-delta-lake-logs.

[27] Microsoft, “EXPLAIN for Synapse Dedicated SQL,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/sql/t-sql/queries/explain-transact-sql?view=azure-sqldw-latest.

[28] Microsoft, “Fabric Data Warehouse performance guidelines,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/guidelines-warehouse-performance.

[29] Microsoft, “In-memory and disk caching,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/caching.

[30] Microsoft, “Parallel Data Warehouse components,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/previous-versions/sql/analytics-platform-system/parallel-data-warehouse-overview?view=aps-pdw-2016-au7.

[31] Microsoft, “Partitioning for Delta tables,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-engineering/delta-lake-partitioning.

[32] Microsoft, “Primary, foreign and unique keys,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/table-constraints.

[33] Microsoft, “Query acceleration frequently asked questions,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/query-acceleration-faq.

[34] Microsoft, “Query acceleration in Fabric Data Warehouse (limited preview),” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/query-acceleration.

[35] Microsoft, “Query insights in Fabric Data Warehouse,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/query-insights.

[36] Microsoft, “Query the Warehouse or SQL analytics endpoint,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/query-warehouse.

[37] Microsoft, “REORG command for Delta Lake tables,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-engineering/delta-lake-reorg.

[38] Microsoft, “Restore in place in Fabric Data Warehouse,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/restore-in-place.

[39] Microsoft, “Result set caching,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/result-set-caching.

[40] Microsoft, “Run Delta table maintenance in Lakehouse,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-engineering/lakehouse-table-maintenance.

[41] Microsoft, “SET SHOWPLAN_XML,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/sql/t-sql/statements/set-showplan-xml-transact-sql?view=fabric.

[42] Microsoft, “SQL analytics endpoint performance considerations,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-engineering/sql-analytics-endpoint-performance.

[43] Microsoft, “Statistics in Fabric Data Warehouse,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/statistics.

[44] Microsoft, “Synapse SQL architecture,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/azure/synapse-analytics/sql/overview-architecture.

[45] Microsoft, “sys.sp_get_table_health_metrics (Transact-SQL),” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/sql/relational-databases/system-stored-procedures/sp-get-table-health-metrics-transact-sql?view=fabric.

[46] Microsoft, “sys.sp_rename in Fabric Warehouse,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/sql/relational-databases/system-stored-procedures/sp-rename-transact-sql?view=fabric.

[47] Microsoft, “T-SQL features in Synapse SQL pools,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/azure/synapse-analytics/sql/overview-features.

[48] Microsoft, “Table maintenance and optimization across Fabric workloads,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/fundamentals/table-maintenance-optimization.

[49] Microsoft, “Tables in Fabric Data Warehouse,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/tables.

[50] Microsoft, “Time travel in Fabric Data Warehouse,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/time-travel.

[51] Microsoft, “Transactions in Fabric Data Warehouse,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/transactions.

[52] Microsoft, “Troubleshoot query blocking,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/troubleshoot-query-blocking.

[53] Microsoft, “Troubleshoot the Warehouse,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/troubleshoot-fabric-data-warehouse.

[54] Microsoft, “Usage reporting in Fabric Data Warehouse,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/usage-reporting.

[55] Microsoft, “Use data clustering with CREATE TABLE AS SELECT,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/tutorial-data-clustering.

[56] Microsoft, “Use Iceberg tables with OneLake,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/onelake/onelake-iceberg-tables.

[57] Microsoft, “V-Order in Fabric Warehouse,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/v-order.

[58] Microsoft, “Warehouse snapshots,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/warehouse-snapshot.

[59] Microsoft, “What is Fabric Data Warehouse?,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/data-warehousing.

[60] Microsoft, “Workload management,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/workload-management.

[61] Microsoft, “Zero-copy table clones,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/clone-table.

[62] Apache Software Foundation, “Apache Parquet dictionary encoding,” Apache Parquet specification. [Online]. Available: https://parquet.apache.org/docs/file-format/data-pages/encodings/#dictionary-encoding-plain_dictionary--2-and-rle_dictionary--8.

[63] Apache Software Foundation, “Apache Parquet file format,” Apache Parquet specification. [Online]. Available: https://parquet.apache.org/docs/file-format/.

[64] Apache Software Foundation, “Apache Parquet metadata,” Apache Parquet specification. [Online]. Available: https://parquet.apache.org/docs/file-format/metadata/.

[65] Apache Software Foundation, “Apache Parquet metadata structures,” Apache Parquet specification. [Online]. Available: https://github.com/apache/parquet-format/blob/master/src/main/thrift/parquet.thrift.

[66] Apache Software Foundation, “Apache Parquet nested encoding,” Apache Parquet specification. [Online]. Available: https://parquet.apache.org/docs/file-format/nestedencoding/.

[67] Apache Software Foundation, “Apache Parquet pages and encodings,” Apache Parquet specification. [Online]. Available: https://parquet.apache.org/docs/file-format/data-pages/.

[68] Apache Software Foundation, “Compression,” Apache Parquet specification. [Online]. Available: https://parquet.apache.org/docs/file-format/data-pages/compression/.

[69] Delta Lake contributors, “Delta Lake deletion vectors,” Delta Lake specification and documentation. [Online]. Available: https://docs.delta.io/delta-deletion-vectors/.

[70] Delta Lake contributors, “Delta Transaction Log Protocol,” Delta Lake specification and documentation. [Online]. Available: https://github.com/delta-io/delta/blob/master/PROTOCOL.md.

[71] Delta Lake contributors, “Universal Format (UniForm),” Delta Lake specification and documentation. [Online]. Available: https://docs.delta.io/delta-uniform/.

[72] Substrait contributors, “About Substrait,” Substrait specification and project resources. [Online]. Available: https://substrait.io/about/.

[73] Substrait contributors, “Substrait Apache License 2.0,” Substrait specification and project resources. [Online]. Available: https://github.com/substrait-io/substrait/blob/main/LICENSE.

[74] Substrait contributors, “Substrait binary serialization,” Substrait specification and project resources. [Online]. Available: https://substrait.io/serialization/binary_serialization/.

[75] Substrait contributors, “Substrait common fields: emit,” Substrait specification and project resources. [Online]. Available: https://substrait.io/relations/common_fields/#emit.

[76] Substrait contributors, “Substrait extensions and function declarations,” Substrait specification and project resources. [Online]. Available: https://substrait.io/extensions/.

[77] Substrait contributors, “Substrait field references,” Substrait specification and project resources. [Online]. Available: https://substrait.io/expressions/field_references/.

[78] Substrait contributors, “Substrait logical relations,” Substrait specification and project resources. [Online]. Available: https://substrait.io/relations/logical_relations/.

[79] Substrait contributors, “Substrait project governance and contributor affiliations,” Substrait specification and project resources. [Online]. Available: https://substrait.io/governance/.

[80] Substrait contributors, “Substrait relational plan format,” Substrait specification and project resources. [Online]. Available: https://substrait.io/.

[81] Databricks, “Dynamic file pruning,” Databricks documentation and engineering blog. [Online]. Available: https://docs.databricks.com/aws/en/optimizations/dynamic-file-pruning.

[82] Databricks, “Faster SQL Queries on Delta Lake with Dynamic File Pruning,” Databricks documentation and engineering blog. [Online]. Available: https://www.databricks.com/blog/2020/04/30/faster-sql-queries-on-delta-lake-with-dynamic-file-pruning.html.

[83] Apache Software Foundation, “PySpark DataFrame.coalesce,” Apache Spark documentation. [Online]. Available: https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.DataFrame.coalesce.html.

[84] Apache Software Foundation, “PySpark DataFrame.repartition,” Apache Spark documentation. [Online]. Available: https://spark.apache.org/docs/latest/api/python/reference/pyspark.sql/api/pyspark.sql.DataFrame.repartition.html.

[85] Microsoft, “Automatic log checkpointing for Fabric Warehouse,” Microsoft Fabric announcements. [Online]. Available: https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/announcing-automatic-log-checkpointing-for-fabric-warehouse/5172079.

[86] C. Cunningham, M. Ketema, and F. Popovic, “Data Warehousing for the AI Era: Deep Dive into High-Performance Analytics and GPU-acceleration, W29,” FabCon Europe, Sept. 30, 2026; supplied slides. [Online]. Available: https://espc.tech/conference/fabcon-europe-2026/programme/.

[87] J. Đoković, I. Zagorac, and F. Santos, “Enterprise Data Warehousing on Fabric: Migrate, Govern, and Scale, TH13,” FabCon Europe, Oct. 1, 2026; supplied slides. [Online]. Available: https://espc.tech/conference/fabcon-europe-2026/programme/enterprise-data-warehousing-on-fabric-migrate-govern-and-scale/.

[88] Microsoft, “Fabric September 2026 Feature Summary,” Microsoft Fabric announcements. [Online]. Available: https://community.fabric.microsoft.com/blog/fbc_fabricupdatesblogs/fabric-september-2026-feature-summary/5325825.

[89] Microsoft, “File-Level write-write conflict detection (Generally Available),” Microsoft Fabric roadmap, planned Q4 2026. [Online]. Available: https://roadmap.fabric.microsoft.com/?product=datawarehouse&releaseItem=b7e155c6-94bf-f011-bbd3-000d3a5b0efa.

[90] Microsoft, “Introducing Microsoft Fabric: Data analytics for the era of AI,” Microsoft Fabric announcements. [Online]. Available: https://azure.microsoft.com/en-us/blog/introducing-microsoft-fabric-data-analytics-for-the-era-of-ai/.

[91] M. Ketema, “Powering modern data analytics in Fabric Data Warehouse, OD813,” Microsoft Build, June 2–3, 2026; supplied transcript and screenshots. [Online]. Available: https://build.microsoft.com/en-US/sessions/OD813?source=sessions.

[92] Microsoft, “Prepare your data for AI innovation with Microsoft Fabric, now generally available,” Microsoft Fabric announcements. [Online]. Available: https://www.microsoft.com/en-us/microsoft-fabric/blog/2023/11/15/prepare-your-data-for-ai-innovation-with-microsoft-fabric-now-generally-available/.

[93] T. Tereshko and J. Podgoetsky, “T16: The Evolution of Fabric Data Warehouse Architecture, Performance, and the Road Ahead,” FabCon Europe, Sept. 29, 2026; supplied slides. [Online]. Available: https://espc.tech/conference/fabcon-europe-2026/programme/corenote-the-evolution-of-fabric-data-warehouse-architecture-performance-and-the-road-ahead/.

[94] Transaction Processing Performance Council, “TPC decision-support benchmarks,” TPC benchmark resources. [Online]. Available: https://www.tpc.org/tpc-bm-categories/decision-support5.asp.

[95] Apache Software Foundation, “Apache Parquet,” Apache Parquet specification. [Online]. Available: https://parquet.apache.org/.

[96] Apache Software Foundation, “Apache Parquet,” Apache Parquet specification. [Online]. Available: https://github.com/apache/parquet-format/blob/master/logo/parquet-logos_3.svg.

[97] Microsoft, “Azure Synapse Analytics icon,” Azure Architecture Center, artwork. [Online]. Available: https://learn.microsoft.com/en-us/azure/architecture/icons/.

[98] Lucide contributors, “Component pictograms,” Lucide Icons. [Online]. Available: https://lucide.dev.

[99] Microsoft, “Custom SQL pools diagram,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/media/custom-sql-pools/custom-sql-pools.svg.

[100] Delta Lake contributors, “Delta Lake,” Delta Lake specification and documentation. [Online]. Available: https://delta.io/.

[101] Microsoft, “Distributed query processing diagram,” Microsoft Learn. [Online]. Available: https://learn.microsoft.com/en-us/fabric/data-warehouse/media/architecture/distributed-query-processing.png.

[102] Microsoft, “Microsoft Fabric icons: Warehouse, Lakehouse and OneLake,” Microsoft Learn, artwork. [Online]. Available: https://learn.microsoft.com/en-us/fabric/fundamentals/icons.

[103] Google, “Google,” official artwork. [Online]. Available: https://www.google.com/.

[104] IBM, “IBM,” IBM Design Language, artwork. [Online]. Available: https://www.ibm.com/design/language/ibm-logos/8-bar/.

[105] Intel, “Intel,” official artwork. [Online]. Available: https://www.intel.com/content/dam/logos/intel-header-logo.svg.

[106] NVIDIA, “NVIDIA,” product resources and artwork. [Online]. Available: https://nvidianews.nvidia.com/multimedia/corporate/nvidia-logos.

[107] NVIDIA, “NVLink interconnect illustration,” product resources and artwork. [Online]. Available: https://www.nvidia.com/en-us/data-center/nvlink/.

[108] Substrait contributors, “Substrait,” Substrait specification and project resources. [Online]. Available: https://substrait.io/img/logo.svg.

The Polaris Papers: Inside the Fabric Data Warehouse

Slides

Benchmarks and query plans

Research papers

Formats and documentation

Fabric Data Warehouse behavior

Supporting references

See you at the next one?

The Polaris Papers: Inside the Fabric Data Warehouse

Get the slides

Slides

Benchmarks and query plans

Research papers

Formats and documentation

Fabric Data Warehouse behavior

Supporting references

See you at the next one?