Back to papers
Biathlon: Harnessing Model Resilience for Accelerating ML Inference Pipelines
Summary: Biathlon, an ML-serving system, exploits model resilience to input perturbations by selecting per-aggregation-feature approximation levels to maximize latency reduction while guaranteeing bounded end-to-end accuracy loss. Evaluated on real pipelines, it achieves 5.3x–16.6x speedups with negligible accuracy drop.
(summarized by gpt-5-mini on Feb 09 2026)
- Paper ID
- 13486
- Venue
- VLDB
- Year
- 2024
- Pagerank
- 4.5911668e-05
- Overall Rank
- 8,080 | 43.79%
- DOI
-
10.14778/3675034.3675052
Incoming Non-self Citations Over Time
Incoming Citations (Sorted by Pagerank)
Showing 3 of 3 citing papers.
Outgoing Citations (Sorted by Pagerank)
Showing 27 of 27 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank |
Cited Paper |
Year |
Venue |
Pagerank |
| 14 |
Online Aggregation |
1997 |
SIGMOD |
0.0010801504 |
| 316 |
NoScope: Optimizing Neural Network Queries over Video at Scale |
2017 |
VLDB |
0.00027988668 |
| 608 |
DeepDB: Learn from Data, not from Queries! |
2020 |
VLDB |
0.00019235898 |
| 943 |
Wander Join: Online Aggregation via Random Walks |
2016 |
SIGMOD |
0.00015145883 |
| 1,204 |
VerdictDB: Universalizing Approximate Query Processing |
2018 |
SIGMOD |
0.00013319541 |
| 1,323 |
Quickr: Lazily Approximating Complex AdHoc Queries in BigData Clusters |
2016 |
SIGMOD |
0.00012601997 |
| 1,874 |
Knowing When You’re Wrong: Building Fast and Reliable Approximate Query Processing Systems |
2014 |
SIGMOD |
0.00010244443 |
| 2,355 |
G-OLA: Generalized On-Line Aggregation for Interactive Analysis on Big Data |
2015 |
SIGMOD |
8.9677847e-05 |
| 2,501 |
DBEst: Revisiting Approximate Query Processing Engines with Machine Learning Models |
2019 |
SIGMOD |
8.6453446e-05 |
| 2,580 |
Sample + Seek: Approximating Aggregates with Distribution Precision Guarantee |
2016 |
SIGMOD |
8.5058814e-05 |
| 2,588 |
Database Learning: Toward a Database that Becomes Smarter Every Time |
2017 |
SIGMOD |
8.4909562e-05 |
| 2,804 |
Extending Relational Query Processing with ML Inference |
2020 |
CIDR |
8.0935487e-05 |
| 2,896 |
Evaluating End-to-End Optimization for Data Analytics Applications in Weld |
2018 |
VLDB |
7.9452051e-05 |
| 3,167 |
Relational Confidence Bounds Are Easy With The Bootstrap* |
2005 |
SIGMOD |
7.4523397e-05 |
| 3,254 |
Query Processing on Tensor Computation Runtimes |
2022 |
VLDB |
7.3161051e-05 |
| 3,331 |
A Demonstration of Willump: A Statistically-Aware End-to-end Optimizer for Machine Learning Inference |
2020 |
VLDB |
7.2131599e-05 |
| 3,407 |
End-to-end Optimization of Machine Learning Prediction Queries |
2022 |
SIGMOD |
7.1295646e-05 |
| 3,842 |
Turbo-Charging Estimate Convergence in DBO |
2009 |
VLDB |
6.7102374e-05 |
| 4,687 |
Serving and Optimizing Machine Learning Workflows on Heterogeneous Infrastructures |
2023 |
VLDB |
5.9986055e-05 |
| 4,748 |
Rafiki: Machine Learning as an Analytics Service System |
2019 |
VLDB |
5.9526539e-05 |
| 5,476 |
Containerized Execution of UDFs: An Experimental Evaluation |
2022 |
VLDB |
5.4866534e-05 |
| 6,247 |
Optimizing In-memory Database Engine for AI-powered On-line Decision Augmentation Using Persistent Memory |
2021 |
VLDB |
5.1389201e-05 |
| 6,339 |
Incremental Computation of Common Windowed Holistic Aggregates |
2016 |
VLDB |
5.1051458e-05 |
| 7,920 |
JoinBoost: Grow Trees Over Normalized Data Using Only SQL |
2023 |
VLDB |
4.6163888e-05 |
| 9,364 |
FEBench: A Benchmark for Real-Time Relational Data Feature Extraction |
2023 |
VLDB |
4.3502487e-05 |
| 9,770 |
Everest: A Top-K Deep Video Analytics System |
2022 |
SIGMOD |
4.2856106e-05 |
| 9,786 |
RALF: Accuracy-Aware Scheduling for Feature Store Maintenance |
2024 |
VLDB |
4.2827012e-05 |
Semantically Similar Papers