Recently, new storage media such as flash memory have been developed very quickly, which brings big challenges to the architecture of computer systems as well as the design of system software. In particular, NAND flash (either SLC- or MLC-based) in the form of solid state disks (SSDs) has been an alternative to traditional magnetic disks, both in the home-user environment and in the enterprise computing environment, due to its shock-resistance, low power consumption, non-volatile, and high I/O speed. The special features of flash memory and other new storage media impose new challenges to traditional data management technologies. As a result, traditional database architectures and algorithms designed for magnetic-disk-based storage fail to utilize new storage media efficiently. Meanwhile, the new characteristics of modern storage media, such as not-in-place update and asymmetric read/write/erase latencies of flash memory, also bring great challenges in optimizing database performance, by using new querying algorithms, indexes, buffer management schemes, and new transaction processing protocols. Consequently, exploiting the characteristics of flash memory and other new storage media has become an important topic of database systems research.

In order to make database systems adapt automatically to the characteristics of flash memory and other new storage media, the data management community needs to rethink traditional underlying storage architecture, query processing algorithms, indexing mechanism, buffer management schemes as well as many traditional issues in magnetic-disk-oriented database systems to adapt to the advances in the underlying storage infrastructure.

The aim of this one-day workshop is to bring together researchers who are interested in optimizing database performance on flash memory or other new storage media based storage infrastructure by designing new data management techniques and tools.

Topics

We welcome papers that address fundamental research issues in this challenging area, with emphasis on new algorithms of flash-based data management. We also encourage papers to report on system level research related to data management on flash-based solid state disks or other storage devices. A number of invited papers will also be solicited. In particular, submissions covering topics from the following non-exclusive list are encouraged:

Ø Design of new storage devices, e.g., flash-based solid state disks

Ø Database architectures for new storage devices

Ø Green-computing and energy-efficient techniques for new storage devices

Ø Cost models and query optimization for new storage devices

Ø Index structures optimized for new storage devices

Ø Buffer management policies for new storage devices

Ø Multi-level storage systems

Ø performance analysis of database workloads on new storage devices

Ø Transaction processing in database systems on new storage devices

Ø Data management issues for hybrid storage architecture

Ø New workloads and benchmark tools for algorithms on new storage devices