Ernest: Efficient Performance Prediction for Large-Scale Advanced Analytics

By Shivaram Venkataraman, Zongheng Yang, M. Franklin, B. Recht, Ion Stoica. Symposium on Networked Systems Design and Implementation, 2016

520 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Optimus: an efficient dynamic resource scheduler for deep learning clusters (2018, methodology): “Ernest [62] also adopts a similar approach to estimate the completion time of data analytics.”
Occupy the cloud: distributed computing for the 99% (2017, background): “While several products such as Databricks and Qubole simplify cluster management, the users still need to explicitly start and terminate clusters, and pick the number and type of instances.”
Chameleon: scalable adaptation of video analytics (2018, methodology): “Several recent systems have also faced an exponentially large configuration search space in their problem domains [12, 19, 31, 34].”
CherryPick: Adaptively Unearthing the Best Cloud Configurations for Big Data Analytics (2017, methodology): “For example, Ernest [37] trains a performance model for machine learning applications with a small number of samples but since its performance model is tightly bound to the particular structure of machine learning jobs, it does not work well for applications such as SQL queries (poor adaptivity).”