Corpus ID: 236428179

A Holistic Analysis of Datacenter Operations: Resource Usage, Energy, and Workload Characterization - Extended Technical Report

@article{Versluis2021AHA,
  title={A Holistic Analysis of Datacenter Operations: Resource Usage, Energy, and Workload Characterization - Extended Technical Report},
  author={Laurens Versluis and Mehmet Cetin and Caspar Greeven and Kristian Bruun Laursen and Damian Podareanu and Valeriu Codreanu and Alexandru Uta and Alexandru Iosup},
  journal={ArXiv},
  year={2021},
  volume={abs/2107.11832}
}
Improving datacenter operations is vital for the digital society. We posit that doing so requires our community to shift, from operational aspects taken in isolation to holistic analysis of datacenter resources, energy, and workloads. In turn, this shift will require new analysis methods, and open-access, FAIR datasets with fine temporal and spatial granularity. We leverage in this work one of the (rare) public datasets providing fine-grained information on datacenter operations. Using it, we… Expand

References

SHOWING 1-10 OF 71 REFERENCES
Correlation-aware virtual machine allocation for energy-efficient datacenters
TLDR
This paper presents a power saving solution for datacenters that especially targets the distinctive characteristics of the scale-out applications and takes into account correlation information of core utilization among virtual machines (VMs) in server consolidation to lower actual peak server utilization. Expand
A Reference Architecture for Datacenter Scheduling: Design, Validation, and Experiments
TLDR
A reference architecture for datacenter schedulers is proposed that follows five design principles: components with clearly distinct responsibilities, grouping of related components where possible, separation of mechanism from policy, scheduling as complex workflow, and hierarchical multi-scheduler structure. Expand
Data Centers in the Cloud: A Large Scale Performance Study
TLDR
A large scale survey of in-production data center servers within a time period that spans two years is conducted, providing in-depth analysis on the time evolution of existing data center demands by providing a holistic characterization of typical data center server workloads, by focusing on their basic resource components. Expand
Statistical Characterization of Business-Critical Workloads Hosted in Cloud Datacenters
  • S. Shen, V. V. Beek, A. Iosup
  • Computer Science
  • 2015 15th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing
  • 2015
TLDR
This study sheds light into the workload of cloud data enters hosting business-critical workloads and collects large-scale and long-term workload traces corresponding to requested and actually used resources in a distributed datacenter servicing business- critical workloads. Expand
Data Center Energy Consumption Modeling: A Survey
TLDR
An in-depth study of the existing literature on data center power modeling, covering more than 200 models, organized in a hierarchical structure with two main branches focusing on hardware-centric and software-centric power models. Expand
DCDB Wintermute: Enabling Online and Holistic Operational Data Analytics on HPC Systems
TLDR
This paper proposes Wintermute, a novel generic framework to enable online ODA on large-scale HPC installations, based on a set of logical abstractions to ease the configuration of models at a large scale and maximize code re-use. Expand
Is Big Data Performance Reproducible in Modern Cloud Networks?
TLDR
It is shown how big-data workloads suffer from significant slowdowns and lack predictability and replicability, even when state-of-the-art experimentation techniques are used, and guidelines for practitioners to reduce the volatility of big data performance are provided. Expand
Towards Resource Disaggregation — Memory Scavenging for Scientific Workloads
TLDR
This work designs a memory scavenging technique that makes unused memory available to applications on other cluster nodes already running other tenants' applications, and implements this technique in MemFSS, an in-memory distributed file system. Expand
Identifying Communication Patterns between Virtual Machines in Software-Defined Data Centers
TLDR
The experimental results demonstrate the capability of the proposed approach to identify interacting VMs, even in a challenging scenario where the traffic patterns are similar in every VM belonging to the same application tier. Expand
On the diversity of cluster workloads and its impact on research results
TLDR
An analysis of the private and HPC cluster traces that spans job characteristics, workload heterogeneity, resource utilization, and failure rates shows that the private cluster workloads, consisting of data analytics jobs expected to be more closely related to the Google workload, display more similarity to the HPC Cluster workloads. Expand
...
1
2
3
4
5
...