Conditional Diffusion for Storage Performance Modeling

Aziz Temirkhanov, Daniil Kopytov, Mikhail Hushchyn

Submission Type: I want my submission to be considered for poster only

Keywords: storage systems, performance modeling, conditional generation, diffusion models

TL;DR: We build on the vendor-agnostic, measurement driven framework, which formulates the task as learning 𝑝 (𝑦 | 𝑥) from measurements and evaluates generative surrogates using distributional metrics.

Confirmation Of Submission Requirements: I submit an abstract. It uses the template provided on the submission page and is no longer than 2 pages.

Submission Number: 322