Dongsheng Yang will present his General Exam on Monday, May 9, 2022 at 2pm via Zoom.

Dongsheng Yang will present his General Exam on Monday, May 9, 2022 at 2pm via Zoom. Zoom link: princeton.zoom.us/j/6605501024 The members of his committee are as follows: Kai Li (adviser), Wyatt Lloyd, Ravi Netravali Abstract: Recent work shows the effectiveness of Machine Learning (ML) to reduce cache miss ratios by making better eviction decisions than heuristics. However, state-of-the-art ML caches require many predictions to make an eviction decision, making them not practical for high-throughput caching systems. We introduce Machine learning At the Tail (MAT), a framework to build efficient ML-based caching systems by integrating with a heuristic algorithm. MAT treats the heuristic algorithm as a “filter” to select high- quality samples to train an ML model and likely candidate objects for evictions. We evaluate MAT on 8 production workloads, spanning storage, in-memory caching, and CDNs. The experiments show that MAT can reduce the number of ML predictions of the state-of-the-art ML-based cache from 63 to 2, reducing the software overhead by 31 times, while achieving similar miss ratios. Link to Reading List: [ https://docs.google.com/document/d/1WeBtRlmeY9Oulz6nhGeFoAYKPb4IC6tzkWmKX0gr... | https://docs.google.com/document/d/1WeBtRlmeY9Oulz6nhGeFoAYKPb4IC6tzkWmKX0gr... ] Everyone is invited to attend the talk, and those faculty wishing to remain for the oral exam following are welcome to do so.

Dongsheng Yang will present his General Exam on Monday, May 9, 2022 at 1:30pm via Zoom. Zoom link: princeton.zoom.us/j/6605501024 The members of his committee are as follows: Kai Li (adviser), Wyatt Lloyd, Ravi Netravali Abstract: Recent work shows the effectiveness of Machine Learning (ML) to reduce cache miss ratios by making better eviction decisions than heuristics. However, state-of-the-art ML caches require many predictions to make an eviction decision, making them not practical for high-throughput caching systems. We introduce Machine learning At the Tail (MAT), a framework to build efficient ML-based caching systems by integrating with a heuristic algorithm. MAT treats the heuristic algorithm as a “filter” to select high- quality samples to train an ML model and likely candidate objects for evictions. We evaluate MAT on 8 production workloads, spanning storage, in-memory caching, and CDNs. The experiments show that MAT can reduce the number of ML predictions of the state-of-the-art ML-based cache from 63 to 2, reducing the software overhead by 31 times, while achieving similar miss ratios. Link to Reading List: [ https://docs.google.com/document/d/1WeBtRlmeY9Oulz6nhGeFoAYKPb4IC6tzkWmKX0gr... | https://docs.google.com/document/d/1WeBtRlmeY9Oulz6nhGeFoAYKPb4IC6tzkWmKX0gr... ] Everyone is invited to attend the talk, and those faculty wishing to remain for the oral exam following are welcome to do so.
participants (1)
-
Nicki Mahler