The ganglia distributed monitoring system: design, implementation, and experience
By Matthew L. Massie, Brent N. Chun, David E. Culler. Parallel Computing, 2004
1,485 citations recorded by the source corpus.
Subfields: Computer Networks and Communications
View Paper: https://www.semanticscholar.org/paper/d7d3e7acd5e408d0a1c0f9d15b4c2e4bdb136447
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Networks and Communications
- Giuseppe Aceto
- Antonio Pescapè
- Kevin Lai
- Vivek S. Pai
- Kyoungsoo Park
- Myeongjae Jeon
- Eytan Adar
- Shivaram Venkataraman
Citations in the corpus
- Tycoon: An implementation of a distributed, market-based resource allocation system (2004, positive): “Our prototype uses a simple centralized soft-state server, but the other components would work just as well with more sophisticated and scalable service location systems (e.g., Ganglia [19] and SWORD [21]).”
- Analysis of Large-Scale Multi-Tenant GPU Clusters for DNN Training Workloads (2019, methodology): “We combine logs from Apache YARN [46], our cluster scheduler, utilization information from Ganglia [33], and logs from each job to perform a systematic analysis of cluster utilization.”
- Multi-tenant GPU Clusters for Deep Learning Workloads: Analysis and Implications (2018, methodology): “We combine logs from Apache YARN [48], our cluster scheduler, utilization information from Ganglia [33], and logs from each job to perform a systematic analysis of cluster utilization.”
- Apollo: Scalable and Coordinated Scheduling for Cloud-Scale Computing (2014, neutral): “For example, it can use a tree hierarchy [20] or a directory service with an eventually consistent gossip protocol [8, 26].”
- Cloud monitoring: A survey (2013, methodology): “Most non-Cloud-specific monitoring systems were already designed to provide extensibility and low intrusiveness, and their extension to Cloud scenarios retained such features [59,60,26,66].”
- CoMon: a mostly-scalable monitoring system for PlanetLab (2006, methodology): “Ganglia does allow site-speci.c metrics to be added to the data it collects, but in our case, metrics such as sliver resource usage are not easily integrated into the node-centric data.”