Workshop on Management of Big Data Systems|
September 21, 2012, San Jose, CA
in conjunction with ICAC 2012
Data is growing at an exponential rate and several systems have emerged to store and analyze such large amounts of data. These systems, termed "Big data systems" are fast evolving Examples include the NoSQL storage systems, Hadoop Map-Reduce, data analytics platforms, search and indexing platforms, and messaging infrastructures. These systems address needs for structured and unstructured data across a wide spectrum of domains such as web, social networks, enterprise, cloud, mobile, sensor networks, multimedia/streaming, cyberphysical and high performance systems, and for multiple application verticals such as biosciences, healthcare, transportation, public sector, energy utilities, oil & gas, and scientific computing.
With increasing scale and complexity, managing these big data systems to cope with failures and performance problems is becoming non-trivial. New resource management and scheduling mechanisms are also needed for such systems, and so are mechanisms for tuning and support from platform layers. Several open source and proprietary solutions have been proposed to address these requirements, with extensive contributions from industry and academia. However, there remain substantial challenges, including those that pertain to such systems' autonomic and self-management capabilities.
The objective of the MBDS workshop is to bring together researchers, practitioners, system administrators, system programmers, and others interested in sharing and presenting their perspectives on the effective management of big data systems. The focus of the workshop is on novel and practical, systems-oriented work. MBDS offers an opportunity for researchers and practitioners from industry, academia, and national labs to showcase the latest advances in this area and to also discuss and identify future directions and challenges in all aspects on autonomic management of big data systems.
Papers are solicited on all aspects of big data management. Specific topics of interest include, but are not limited, to the following:
- Autonomic and self-managing techniques for big data management
- Application-level resource management and scheduling mechanisms
- System tuning/auto-tuning and configuration management
- Performance management, fault management, power management
- Scalability challenges
- Complexity challenges, as for composite, cross-tier systems with multiple control loops
- Unified management of 'data in motion' and 'data at rest'
- Dealing with both structured and unstructured data
- Monitoring, diagnosis, and automated behaviour detection
- System-level principles and support for resource management
- Holistic management across hardware and software
- Implications of emerging hardware technologies such as non-volatile memory
- Domain specific challenges in web, cloud, social networks, mobile, sensor networks, streaming analytics, cyber-physical systems
- System building and experience papers for specific industry verticals