Kodiak: Leveraging Materialized Views For Very Low-Latency Analytics Over High-Dimensional Web-Scale Data
Summary: Kodiak delivers low-latency, high-dimensional analytics over petabyte-scale ad events by incrementally maintaining thousands of partitioned, replicated materialized views and automatically selecting views per query. Production results: 3+ PB, 8 ms median latency, and far lower resource use than big-data systems. (summarized by gpt-5.6-luna on Jul 24 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Shaosu Liu (Turn, Inc.)
- 2. Bin Song (Turn, Inc.)
- 3. Sriharsha Gangam (Turn, Inc.)
- 4. Lawrence Lo (Turn, Inc.)
- 5. Khaled Elmeleegy (Turn, Inc.)
BibTeX Citation
@article{liu_vldb16,
title = {{Kodiak: Leveraging Materialized Views For Very Low-Latency Analytics Over High-Dimensional Web-Scale Data}},
author = {Liu, Shaosu and Song, Bin and Gangam, Sriharsha and Lo, Lawrence and Elmeleegy, Khaled},
journal = {PVLDB},
series = {{VLDB} '16},
volume = {9},
number = {13},
pages = {1269--1280},
doi = {10.14778/3007263.3007270},
url = {https://doi.org/10.14778/3007263.3007270},
year = {2016}
}
Incoming Citations (Sorted by Pagerank)
Showing 3 of 3 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 1,765 | Selecting Subexpressions to Materialize at Datacenter Scale | 2018 | VLDB | 9.8079546e-05 |
| 2,747 | Moment-Based Quantile Sketches for Efficient High Cardinality Aggregation Queries | 2018 | VLDB | 8.1711208e-05 |
| 3,605 | Computation Reuse in Analytics Job Service at Microsoft | 2018 | SIGMOD | 7.2640711e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 17 of 17 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 11,609 | High-dimensional Data Cubes | 2022 | VLDB |
| 2 | 7,551 | Enabling Efficient and General Subpopulation Analytics in Multidimensional Data Streams | 2022 | VLDB |
| 3 | 10,672 | Optimizing Block Skipping for High-Dimensional Data with Learned Adaptive Curve | 2025 | SIGMOD |
| 4 | 3,753 | High-Dimensional OLAP: A Minimal Cubing Approach | 2004 | VLDB |
| 5 | 11,399 | QaaD (Query-as-a-Data): Scalable Execution of Massive Number of Small Queries in Spark | 2023 | SIGMOD |
| 6 | 3,082 | Analytics in Motion: High Performance Event-Processing AND Real-Time Analytics in the Same Database | 2015 | SIGMOD |
| 7 | 2,094 | Tupleware: "Big" Data, Big Analytics, Small Clusters | 2015 | CIDR |
| 8 | 7,899 | Modularis: Modular Relational Analytics over Heterogeneous Distributed Platforms | 2021 | VLDB |
| 9 | 8,202 | Cubrick: Indexing Millions of Records per Second for Interactive Analytics | 2016 | VLDB |
| 10 | 9,640 | Supporting Scalable Analytics with Latency Constraints | 2015 | VLDB |