Original title: Storage-Centric System Designs for Enabling Fast, Efficient, and Low-Cost Genomic and Metagenomic Analyses
Nika Mansouri Ghiasi · August 2026
Studying DNA helps doctors treat diseases and keeps food safe. However, genetic data is enormous. Computers struggle to analyze it quickly because of two big bottlenecks. First, moving huge piles of data from storage to computer processors wastes energy and slows everything down. Second, the data is kept compressed to save space, so computers must spend extra time decompressing and formatting it before doing any real work. To solve this, researchers designed specialized storage-centric systems. Instead of moving all the raw files to separate processing units, these systems run analyses directly inside the storage hardware itself. They also allow data to stay highly compressed while still offering fast access. The researchers found that doing computation inside storage tackles both bottlenecks at once. It significantly improves overall system performance, boosts energy efficiency, and lowers operational costs for large-scale genetic analysis.
Imagine you need to find a specific recipe in a huge cookbook. Normally you take the whole book to the kitchen, flip through pages, and then cook. But the book is heavy, and carrying it takes time. Now suppose you could read the recipe right on the shelf without moving the book. In this paper, the storage system is like the shelf, and the analysis is like reading the recipe. By processing data where it is stored, you avoid moving the entire dataset, saving time and energy.
Genomic analyses are critical for precision medicine, outbreak detection, and food safety, but moving huge amounts of data between storage and processors slows everything down and wastes energy. By performing analysis directly inside the storage system, this work cuts data movement and preparation overheads, making large-scale genomic studies faster, cheaper, and more practical for routine use.
The abstract does not specify the scale of evaluation or compare against all existing methods. It focuses on storage-centric designs but does not address other bottlenecks like algorithmic complexity or hardware cost of specialized storage. The results are demonstrated for genomic and metagenomic analyses, but applicability to other data-intensive domains is suggested, not proven.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
Storage-Centric System Designs for Enabling Fast, Efficient, and Low-Cost Genomic and Metagenomic Analyses — Nika Mansouri Ghiasi
Abstract licence: CC0
Explanation generated by Synthesis from the abstract — it is not the original paper. This is a preprint, so it hasn't necessarily been peer reviewed.