Introduction | Topics | Submission | Deadlines | Organization | Program | Registration
Cloud computing promises unlimited, cost-effective and agile computing resources for users. However, this new computing paradigm also poses a unique set of challenges to both cloud providers and users. On the one hand, cloud providers need to ensure that resources being provided are highly available and deliver high performance, while optimizing cloud infrastructure to reduce their operational costs. On the other hand, cloud users need to ensure that their applications receive the best performance from the cloud, while maintaining their budgetary constraints and the terms of any Service Level Agreements (SLAs) they have with their cloud providers.
Given the scale of cloud deployment, systematic analytical approaches are critically needed to provide insights to both providers and users to achieve their respective goals. For instance, cloud providers need to constantly be aware of the running status and/or anomalies in functionality from their cloud, to be able to quickly fix any issues that may arise, to adjust physical resource allocations to ensure that their customers get best performance, or plan which services to offer to get the best return on investment. Similarly, cloud users need to understand the workload to be deployed into the cloud, plan the deployment in a cost-effective way, or ascertain the flexibility and service quality provided by different cloud environments and use this to decide their deployment strategy. Analytics can play a pivotal role in all these scenarios. By gathering insights from the large amount of data from the cloud, both cloud providers and consumers can develop analytical approaches to achieving their respective objectives in spite of the scale that clouds provide.
The purpose of this workshop is to provide a forum for researchers in the related fields to exchange ideas, and share their experiences in developing analytics to better deploy, operate and use the cloud. Specifically, we seek and wish to foster research contributions that draw on statistical analysis, analytical modeling, and machine learning to develop novel solutions in this problem area.
IWCA'14 will be held in conjunction with the International Conference on Cloud Engineering (IC2E) 2014.
Paper submission due: December 1, 2013
Notification of acceptance: January 3, 2014
Final camera-ready papers due: January 17, 2014
Workshop date: March 11, 2014 (held in conjunction with IC2E 2014)
Topics of interest include, but are not limited to, the following:
• Cloud workload measurement and analysis
• Workload behavior modeling
• Analytics for application deployment in cloud
• Performance modeling of cloud applications
• Cloud performance benchmarking
• Resource utilization optimization
• Tracing and problem identification in cloud systems
• Log and monitoring data analysis
• Problem diagnosis and troubleshooting
• Security and intrusion detection
• Reliability engineering, fault management, and disaster recovery
• Design and implementation of analytics systems
• Business optimization in cloud operations
The IWCA workshop invites authors to submit original and unpublished work. Papers should not exceed 6 pages in IEEE style (single-spaced 2-column text using 10-point size type on A4 paper). Authors should submit a PostScript (level 2) or PDF file that will print on a PostScript printer.
• IWCA'14 paper submission site
• All selected papers will be peer-reviewed.
• For each accepted paper, at least one author is required to register and present the paper at the workshop.
• All accepted papers will be published with IEEE Xplore.
• We will submit all accepted workshop papers for possible publication in a special issue of the International Journal on Big Data Intelligence.
Shu Tao (IBM T J Watson Research)
Rich Wolski (UCSB)
Rahul Singh (IBM T J Watson Research)
Theophilus Benson (Duke University)
Lydia Chen (IBM Zurich)
Yanpei Chen (Cloudera)
Yuan Chen (HP Labs)
David Irwin (UMass, Amherst)
Thilo Kielmann (VU University, Amsterdam)
Ningfang Mi (Northeastern University)
Lavanya Ramakrishnan (Lawrence Berkeley National Lab)
Prashant Shenoy (UMass, Amherst)
Christopher Charles Stewart (Ohio State University)
Evgenia Smirni (William and Mary)
Chunqiang Tang (Facebook)
Jon Weissman (University of Minnesota)
Timothy Wood (George Washington University)
9:00-10am Keynote: Lessons Learned from Hosting Scientific Research Applications in the Cloud
Dennis Gannon, Director of Cloud Research Strategy at Microsoft Research
Abstract: Microsoft Research is in its third year of making grants of Windows Azure Cloud resources to the academic community. In that period we have made nearly 200 awards. Also in that time the Azure cloud has evolved from a pure PaaS platform to include IaaS. It now hosts many new services including many of our commercial on-line products and special events such as live streaming video of the Sochi Olympics. This talk will focus on the experience and progress made by the academic researchers using Azure. The research domains for these projects include computer science, environmental and atmospheric science, life sciences and urban science. We will describe what worked well and what failed completely. Early on the projects focused on traditional scalable computing, but the more interesting current work involves machine learning and data analytics, integrating data streams and hosting services and virtual machines to support communities.
10:00-11:00am Session 1
How will your workload look like in 6 years?
Analyzing Wikimedia workload
Ahmed Ali-Eldin (Umea university), Ali Rezaie (Umea university), Stanislav Razroev (Umea university), Amardeep (Umea University), Sara Sjostedt-de Luna (Umea University), Oleg Seleznjev (Umea University), Johan Tordsson (Umea University), Erik Elmroth (Umea University)
Data Loss: An Empirical Analysis in Search of
Best Practices for Prevention
Lance Fiondella (University of Massachusetts, Dartmouth), Rehab El-Kharboutly (Quinnipiac University), Swapna S. Gokhale (University of Connecticut)
11:00 – 11:15am Break
11:15am-12:15pm Session 2
Live Data Migration For Reducing SLA Violations
In Multi-tiered Storage Systems
Jianzhe Tai (Northeastern University), Bo Sheng (University of Massachusetts Boston), Yi Yao (Northeastern University), Ningfang Mi (Northeastern University)
Parallel Hierarchical Affinity Propagation with
Dillon Mark Rose (Florida Institute of Technology), Jean Michel Rouly (George Mason University), Rana Haber (Florida Institute of Technology), Nenad Mijatovic (Florida Institute of Technology), Adrian M. Peter (Florida Institute of Technology)
1:30-2:30pm Panel - Topic: The Convergence of Cloud, Big Data and Mobile: What could happen in 5 years?
Panelists: Azer Bestavros (Boston
University), Evgenia Smirnik (William & Mary), David Irwin (UMass, Amherst), Tim
Wood (George Washington University). Moderator: Rich Wolski (UCSB)
2:30 – 3:30pm Session 3
Virtual-to-Physical Mapping Inference in
Virtualized Cloud Environments
Yang Song (IBM Research), Ramani Routray (IBM Research), Rakesh Jain (IBM Research)
VMDedup: Memory De-duplication in Hypervisor
Furquan Shaikh (University of Illinois at Urbana-Champaign), Fangzhou Yao (University of Illinois at Urbana-Champaign), Indranil Gupta (University of Illinois at Urbana-Champaign), Roy H. Campbell (University of Illinois at Urbana-Champaign)
3:45 – 4:45pm Session 4
Topology Discovery & Service Classification for
Jinho Hwang (The George Washington University), Guyue Liu (The George Washington University), Timothy Wood (The George Washington University), Sai Zheng (IBM Research), Frederick Wu (IBM Research)
Experiences with Cost and Utility Trade-offs on
IaaS Clouds, Grids, and On-Premise Resources
Tiziano Passerini (Emory University), Jaroslaw Slawinski (Emory University), Umberto Villa (Emory University), Vaidy Sunderam (Emory University)
4:45 – 5:00pm Town Hall Discussion