Title :
Intercloud Networks Performance Analysis
Author :
McGrath, M. Garrett ; Raycroft, Patrick ; Brenner, Paul R.
Author_Institution :
Cntr. for Res. Comput., Univ. of Notre Dame, Notre Dame, IN, USA
Abstract :
Continued growth in cloud IaaS adoption motivates greater transparency into IaaS network performance to effectively leverage intra and intercloud service and data migration needs of a global consumer base. In this work, we produce a baseline set of network performance measurements for cloud WANs both internal to individual IaaS provider´s globally distributed datacenters and in between those provider´s private infrastructures. Standard virtual machine instances provided by major IaaS vendors were utilized along with traditional open source operating systems and latency/bandwidth testing toolsets. The measurements provide insights into bandwidth and latency bottlenecks relative to global data transit between commercial production cloud resources. This data provides network engineers, IaaS/PaaS/SaaS providers, cloud service consumers, and intercloud testbed developers essential reference data upon which to set performance and QoS expectations and propose next generation cloud network infrastructure and software enhancements. Further, the evolution of baseline network metrics for cloud services helps provide uniformity and clarity to IaaS vendor performance and capability comparisons.
Keywords :
cloud computing; operating systems (computers); public domain software; quality of service; IaaS vendor performance; PaaS providers; QoS expectations; SaaS providers; bandwidth testing toolsets; baseline network metrics; baseline set; capability comparisons; cloud service consumers; commercial production cloud resources; data migration; global consumer base; globally distributed datacenters; intercloud networks performance analysis; intercloud service; intercloud testbed developers essential reference data; internal datacenters; intracloud service; latency toolsets; network engineers; network performance measurements; next generation cloud network infrastructure; private infrastructures; software enhancements; traditional open source operating systems; Bandwidth; Cloud computing; Data transfer; Standards; Testing; Virtual machining; Wide area networks; IaaS; Intercloud; TCP; WAN; bandwidth; data transit; latency; network performance; testbed; virtual machines;
Conference_Titel :
Cloud Engineering (IC2E), 2015 IEEE International Conference on
Conference_Location :
Tempe, AZ
DOI :
10.1109/IC2E.2015.85