QoS Assurance in the Cloud – QAC’2015
The first International Workshop on
Quality of Service Assurance in the Cloud
In conjunction with the 7th IEEE International Conference on Cloud Computing Technology and Science (CloudCom2015)
November 30 - December 3 2015, Vancouver Canada
Cloud computing offers a computational model where network, storage, and compute domains can be abstracted into virtual resources offered on demand. While these abstractions offer opportunities to significantly simplify the management and optimize the use of resources, they impose additional challenges on assuring the quality of service (QoS).
In the context of multi-tenancy, the same physical resources are shared by different users that do not necessarily share the same service level agreements; hence the QoS distinction has to be enforced at a finer grain. Moreover, certain attributes of the QoS may contradict with others, for instance, high availability favors replication and distributed redundancy while energy efficiency favors de-duplication and consolidation. The multi-layered architecture of the distributed cloud requires not only a thorough orchestration between layers to ensure the required QoS, but also makes the root cause analysis even more complicated in case of the SLA violation. Continuous changes and migrations of the resources allocated at all layers of abstraction pose further challenges to detecting problems, identifying their location and reasoning about causality.
The aim of this workshop is to raise awareness of the challenges of the QoS assurance in the cloud, and bring together practitioners and researchers to discuss potential solutions.
The workshop solicits high-quality papers authored by both industry and academic contributors that address at least one of the following topics:
¨ Security assurance and regulations compliance in the cloud
¨ Multi-level isolation orchestration for the cloud tenants
¨ High Availability and fault tolerance approaches and techniques in the cloud
¨ Dynamic performance assessment (compute, storage and networking) and prediction
¨ Telemetry and smart metering in the cloud
¨ Carbon Footprint quantification approaches for cloud hosted applications
¨ Requirement modeling and specification for the QoS
¨ Orchestration and refinement for cloud QoS policies
¨ Dynamic verification and validation of cloud service quality
¨ Automated generation of multi-layer QoS test cases
¨ Autonomic approaches for assurance of cloud service quality
¨ Real-time analytics integration with QoS control loops
¨ Evaluations of cloud QoS impact on application or service QoE
¨ Automatic generation of workflows related to assuring QoS in cloud
¨ Machine learning and data mining techniques for optimizing QoS assurance processes
The workshop papers should be no more than 6 pages (including references) in IEEE double column format. Papers should be submitted electronically via the workshop submission site. For an accepted paper, at least one author must register and attend the workshop.
The workshop proceedings will be included in the conference proceedings that will be published in the IEEE Xplore library.
August 15 (extended) closed
Authors Notification date: September 1
Camera ready version: September 15
Ali Kanso, Ericsson Research, Canada: firstname.lastname@example.org
Catalin Meirosu, Ericsson Research, Sweden: catalin.meirosu@ericsson .com
Seetharami R. Seelam, IBM T. J. Watson Research Center, USA
Michela D'Errico, Hewlett-Packard Laboratories, UK
Khoder Shamy, BlackBerry Limited, Canada
Ewa Deelman, University of Southern California, USA
Renato Figueiredo, University of Florida, USA
Georges Khalil, Google Inc., Canada
Rolf Stadler, KTH Royal Institute of Technology, Sweden
Radu Prodan, University of Innsbruck, Austria
Mohamed Cheriet, Ecole de Technology Superieure, Canada
Sanjeev Mehrotra, Microsoft Research, USA
Gunter Mussbacher, McGill University, Canada
Pietro Colombo, Insubria University, Italy
Andreas Johnsson, Ericsson Research, Sweden
Ferhat Khendek, Concordia University, Canada
Devin Parrish, Mirantis, USA
Daniel Migault, Ericsson Research, Canada
Erik Elmroth, Umeå University, Sweden
Stephan Schmid, Technical University of Berlin, Germany
Abdelouahed Gherbi, Ecole de Technology Superieure, Canada
Rami Bahsoon, Birmingham University, United Kingdom
Simin Nadjm-Tehrani, University of Linköping, Sweden
Geoffrey Lefebvre, Coho Data, Canada
Tereza Carvalho, University of Sao Paulo, Brazil
Our invited speaker is Seetharami R. SeelamSeelam. Dr. Seelam is a Research Staff Member in the IBM Distributed Systems group at the Thomas J. Watson Research Center in New York.
Dr. Seelam received his Ph.D. degree in computer science from University of Texas at El Paso in 2006. He subsequently joined IBM Research Center, where he works on next generation cloud (polyglot) middleware, container clouds, data centric computing, hybrid systems, workload optimized systems, parallel and distributed systems, operating systems, resource management and system performance evaluation for commercial and scientific workloads. He worked on IBM Blue Gene and DARPA High Productivity Computing System (HPCS) projects. Currently he leads a team of researchers to reinvent enterprise middleware for the new millennium by uniquely combining system, middleware and workload patterns.
He published in international conferences and journals and received two best paper awards and an outstanding paper award from International Parallel and Distributed Symposium (IPDPS), International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD), and International Conference on High Performance Computing and Communications (HPCC), respectively. He has nine patents granted; more than 20 patents filed with USPTO and received four Plateau awards in recognition of his innovations to IBM. He received two innovation achievement awards for high value patent applications. He received two outstanding technical accomplishment award (OTAA) from IBM for his work on Java software stack optimization of various products on POWER platform and for his work on the Operating System design for HPCS. His work resulted in more than a USD 100 Million impact on IBM's revenue.
Dr. Seelam will present an overview of the IBM container Cloud, focusing on the PaaS and Autoscaling challenges in order to satisfy the SLA requirements. He will present the Autoscaling solution; Principles and Architecture. Finally, Dr. Seelam will conclude the talk discussing the feasibility of having Autoscaling as a service, and the future challenges.
13:30 – 14:10
Keynote Presentation: The IBM Container Cloud: Principles and Practices in Assuring the Quality of Service
Dr. Seetharami R. SeelamSeelam, Research Staff Member at IBM T.J.Watson. New York.
14:10 – 14:35
Minimizing Latency of Real-Time Container Cloud for Software Radio Access Networks
Chen-Ning Mao, Mu-Han Huang, Satyajit Padhy, Shu-Ting Wang, Wu-Chun Chung, Yeh-Ching Chung, and Cheng-Hsin Hsu
14:35 – 15:00
Simulating High Availability Scenarios in Cloud Data Centers: A Closer Look
Mohamed Abu Sharkh, Abdallah Shami, , Abdelkader Ouda, Peter Ohlen and Ali kanso
15:00 – 15:30
15:30 – 15:55
A Dynamic Bandwidth Prediction & Provisioning Scheme in Cloud Networks
15:55 – 16:20
A Formal Approach for QoS Assurance in the Cloud
Parisa Heidari, Hanifa Boucheneb and Abdallah Shami
16:20 – 16:45
Temperature Distribution Prediction in Data Centers for Decreasing Power Consumption by Machine Learning
Yuya Tarutani, Kazuyuki Hashimoto, Go Hasegawa, Yutaka Nakamura, Takumi Tamura, Kazuhiro Matsuda, and Morito Matsuoka
16:45 – 17:00
Workshop closing notes