# KML: Using Machine Learning to Improve Storage Systems

arXiv: https://arxiv.org/abs/2111.11554 · HTML: https://mlsanity.com/paper/2111.11554/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 97 references that are in the map.
- First posted on arXiv 22 Nov 2021.

## Builds on (most cited first, 13 of 13)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activations](https://mlsanity.com/paper/1609.07061/) (2016)
- [Tune: A Research Platform for Distributed Model Selection and Training](https://mlsanity.com/paper/1807.05118/) (2018)
- [Deep Learning with Limited Numerical Precision](https://mlsanity.com/paper/1502.02551/) (2015)
- [Neural Additive Models: Interpretable Machine Learning with Neural Nets](https://mlsanity.com/paper/2004.13912/) (2020)
- [The Case for Learned Index Structures](https://mlsanity.com/paper/1712.01208/) (2017)
- [Fixed Point Quantization of Deep Convolutional Networks](https://mlsanity.com/paper/1511.06393/) (2015)
- [Explanation Methods in Deep Learning: Users, Values, Concerns and Challenges](https://mlsanity.com/paper/1803.07517/) (2018)
- [Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications](https://mlsanity.com/paper/2003.07631/) (2020)
- [High-Accuracy Low-Precision Training](https://mlsanity.com/paper/1803.03383/) (2018)
- [Deep Convolutional Neural Network Inference with Floating-point Weights and Fixed-point Activations](https://mlsanity.com/paper/1703.03073/) (2017)
- [Running Neural Networks on the NIC](https://mlsanity.com/paper/2009.02353/) (2020)
- [A Learning-based Approach Towards Automated Tuning of SSD Configurations](https://mlsanity.com/paper/2110.08685/) (2021)

## Abstract (arXiv)

Operating systems include many heuristic algorithms designed to improve overall storage performance and throughput. Because such heuristics cannot work well for all conditions and workloads, system designers resorted to exposing numerous tunable parameters to users -- thus burdening users with continually optimizing their own storage systems and applications. Storage systems are usually responsible for most latency in I/O-heavy applications, so even a small latency improvement can be significant. Machine learning (ML) techniques promise to learn patterns, generalize from them, and enable optimal solutions that adapt to changing workloads. We propose that ML solutions become a first-class component in OSs and replace manual heuristics to optimize storage systems dynamically. In this paper, we describe our proposed ML architecture, called KML. We developed a prototype KML architecture and applied it to two case studies: optimizing readahead and NFS read-size values. Our experiments show that KML consumes less than 4KB of dynamic kernel memory, has a CPU overhead smaller than 0.2%, and yet can learn patterns and improve I/O throughput by as much as 2.3x and 15x for two case studies -- even for complex, never-seen-before, concurrently running mixed workloads on different storage devices.