WarpLDA: a Cache Efficient O(1) Algorithm for Latent Dirichlet Allocation

Summary: WarpLDA is a cache-aware O(1) per-token LDA that analyzes per-document memory access to maximize L3 cache locality. Achieves 5–15× speedups over LightLDA with 11B tokens/s throughput, enabling a million topics on 639M documents in five hours. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID: 11363
Venue: VLDB
Year: 2016
Pagerank: 5.2415551e-05
Overall Rank: 6,014 | 58.17%
DOI: -

Incoming Non-self Citations Over Time

Authors

Incoming Citations (Sorted by Pagerank)

Showing 2 of 2 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
11,795	LDA*: A Robust and Large-scale Topic Modeling System	2017	VLDB	4.1945683e-05
13,328	Scalable Training of Hierarchical Topic Models	2018	VLDB	-

Outgoing Citations (Sorted by Pagerank)

Showing 0 of 0 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank

Semantically Similar Papers

Overall Rank	Paper	Year	Venue	Pagerank
8,150	Parallelism-Optimizing Data Placement for Faster Data-Parallel Computations	2023	VLDB	4.5746638e-05
4,761	Efficient Graph Summarization using Weighted LSH at Billion-Scale	2021	SIGMOD	5.9404527e-05
3,323	CRD: Fast Co-clustering on Large Datasets Utilizing Sampling-Based Matrix Decomposition	2008	SIGMOD	7.2224696e-05
11,466	Fast Density-Peaks Clustering: Multicore-based Parallelization Approach	2021	SIGMOD	4.1945683e-05
10,266	Near-Duplicate Text Alignment under Weighted Jaccard Similarity	2026	VLDB	4.1945683e-05
428	Latent Semantic Indexing: A Probabilistic Analysis	1998	PODS	0.00023512226
1,967	Compressed Linear Algebra for Large-Scale Machine Learning	2016	VLDB	9.9131712e-05
328	An Architecture for Parallel Topic Models	2010	VLDB	0.0002728514
13,328	Scalable Training of Hierarchical Topic Models	2018	VLDB	-
11,795	LDA*: A Robust and Large-scale Topic Modeling System	2017	VLDB	4.1945683e-05