The Forth International Workshop on
  Efficient Data Center System (EDCS)

  Held in conjunction with ISCA-2016

June 19,2016
Seoul, Korea


Cloud data centers today are becoming overwhelmed by the vast amount of raw data generated from various distributed data sources. The challenge for handling these datasets in the cloud is increasing rapidly due to the time-consuming bulk data transfer and the escalating power capacity needs for analyzing them in a single location. As the internet of Things (IoT) applications and cyber-physical systems (CPS) rapidly grow, efficient data hosting and low-latency analytics on ubiquitous datasets have become increasingly important. According to a recent study by Gartner, transferring all these distributed datasets to a central location for processing will not be technically and economically viable in the ubiquitous big data era Therefore, it is timely to rethink the architecture and resource management of existing data center systems that places heavy emphasis on building central warehouse-scale server clusters.

The Forth EDCS Workshop will be held in conjunction with the 43rd ACM/IEEE International Symposium on Computer Architecture (ISCA-2016) in Seoul, Korea. The EDCS Workshop Series cover a wide range of topics in data center design. The 1st round of EDCS mainly focused on micro-server systems; the 2nd round of EDCS centered on the issues of infrastructure and deployment; the 3rd round of EDCS looks at workload analysis and architecture exploration. This year the theme will be data center design for ubiquitous big data.

 

EDCS+ASBD:

This year EDCS occurs jointly with ASBD (Architectures and Systems for Big Data Workshop) in a single-track multi-session program. The two workshops will share the meeting room and keynotes etc.

(Check our Technical Program at: http://acs.ict.ac.cn/asbd2016/ )

 

Accepted Papers :

[1] An Evaluation of Contemporary Heterogeneous Computing Platforms for Data Intensive Applications

[2] OpenStack and Docker Comparison for Scientific Workflow w.r.t. Execution and Energy

[3] A Machine Learning-based Approach to Live Migration Modeling

 

The goal of this workshop is to bring together academic researchers and industry practitioners to share their insights on emerging IoT workloads, exchange their understandings of the design challenges for managing the upcoming ubiquitous big data, and discuss the possible architectural innovations for the next-generation data center and server systems. We will continue the trend of making EDCS an inclusive and interactive event that features invited keynotes and paper presentations. We especially encourage submissions that look forward to future data center systems and technologies in the IoT big data era. However, all topics related to efficient data center designs are welcome:

  • Analysis for understanding ubiquitous big data
  • Methodologies for accelerating data processing
  • Interactions between IoT devices and the cloud
  • Edge computing and modular data centers
  • Architectural support for QoS and QoE
  • Power-efficient system and management
  • Design for scalability and sustainability
  • Cost-effective data center architecture
  • Secure and dependable cloud systems
  • New exploration inspired by emerging techniques
  •  

Important Dates:
  • Abstract Deadline: April 15, 2016
  • Full Paper: April 22, 2016
  • Notification: May 06, 2016
  •  

    Workshop Chair:

    Chao Li
    Assistant Professor, Shanghai Jiao Tong University, China

     

    Steering Committee:

  • Rui Hou
    Associate Professor, Institute of Computing Technology, Chinese Academy of Sciences, China
  • Jian Li
    Research Program Director and Chief Architect, Huawei Technologies, USA
  • Michael C. Huang
    Associate Professor, University of Rochester, USA
  • Jichuan Chang, Google
  • Mike Ferdman, Stony Brook University
  • Boris Grot, University of Edinburgh
  • Inigo Goiri, Microsoft Research
  • Yang Hu, University of Florida
  • Tao Li, University of Florida and NSF
  • Xiaoyi Lu, Ohio State University
  • Xuehai Qian, University of Sothern California
  • Shaolei Ren, University of California Riverside
  • Di Wang, Microsoft Research
  • Jun Yao, Huawei

We welcome short papers on emerging ideas and work-in-progress achievements. Papers must be submitted in PDF format and should contain a maximum of 4 pages of single-spaced text (using the IEEE two-column format), not including references. Manuscript submissions are managed by easyChair.  Please use the following link to submit your paper: https://easychair.org/conferences/?conf=edcs2016

Main Program (The Joint Workshop of ASBD and EDCS):

http://acs.ict.ac.cn/asbd2016/