@inproceedings{jyothi2016morpheus,
author = {Jyothi, Sangeetha Abdu and Curino, Carlo and Menache, Ishai and Matthur Narayanamurthy, Shravan and Tumanov, Alexey and Yaniv, Jonathan and Mavlyutov, Ruslan and Goiri, Íñigo and Krishnan, Subru and Kulkarni, Janardhan (Jana) and Rao, Sriram},
title = {Morpheus: Towards Automated SLOs for Enterprise Clusters},
booktitle = {2016 International Symposium on Operating Systems Design and Implementation (OSDI)},
year = {2016},
month = {November},
abstract = {Modern resource management frameworks for large-scale analytics leave unresolved the problematic tension between high cluster utilization and job’s performance predictability—respectively coveted by operators and users. We address this in Morpheus, a new system that: 1) codifies implicit user expectations as explicit Service Level Objectives (SLOs), inferred from historical data, 2) enforces SLOs using novel scheduling techniques that isolate jobs from sharing-induced performance variability, and 3) mitigates inherent performance variance (e.g., due to failures) by means of dynamic reprovisioning of jobs. We validate these ideas against production traces from a 50k node cluster, and show that Morpheus can lower the number of deadline violations by 5x to 13x, while retaining cluster-utilization, and lowering cluster footprint by 14% to 28%. We demonstrate the scalability and practicality of our implementation by deploying Morpheus on a 2700-node cluster and running it against production-derived workloads.

 },
url = {http://approjects.co.za/?big=en-us/research/publication/morpheus-towards-automated-slos-enterprise-clusters/},
edition = {International Symposium on Operating Systems Design and Implementation (OSDI)},
}