UW-Madison Logo

The ADvanced Systems Laboratory (ADSL)
Publication abstract

:: Home
:: Projects
:: People
:: Pictures
:: Publications
:: Software
:: Sponsors
:: Internal

Semantically-smart Disk Systems

Muthian Sivathanu, Department of Computer Sciences, University of Wisconsin-Madison

Abstract:

Robust and efficient storage of data is a prerequisite of current and future computer systems. To keep pace with both rapid changes in technology as well as increasing demands from applications and users alike, storage systems must evolve in new and interesting ways.

Unfortunately, storage systems today have a problem: the range of functionality they can provide is fundamentally limited, despite the presence of significant processing power within them. The main reason for this limitation is that storage systems communicate with the outside world through a narrow block-based interface today, and therefore lack higher-level ``semantic'' understanding about how they are being used.

This thesis proposes a solution to this fundamental problem. It presents a new class of storage systems called ``semantically-smart disk systems'' (SDS's); such disk systems are capable of providing entirely new classes of functionality by exploiting information about the system above (e.g., a file system or a database management system). An SDS does so by carefully monitoring the low-level stream of block reads and block writes that a storage system normally sees, and then inferring higher-level behaviors of the system above. Importantly, an SDS does so without any changes to the existing block-level storage interface, taking a pragmatic approach that enables ready deployment in existing computing environments.

In this thesis, we present a variety of techniques used by an SDS to track semantic information underneath modern file systems, demonstrating how to transform an I/O request stream into a source of useful high-level information for the underlying disk system. We also demonstrate the utility of semantic information within the disk system by presenting new improvements to the availability, security, and performance of storage. For example, we have built a storage system that exhibits much better availability under multiple failures by keeping semantically-meaningful data available. In another case study, we show that semantic knowledge within the storage system can enable reliable secure deletion of data. Such innovations are impossible to implement in the current storage infrastructure, but become possible with the acquisition and careful use of semantic information. Finally, we present a new logic framework for reasoning about file systems and their interaction with storage systems, and use this logic to prove properties about inference within a semantically-smart disk system.

Available as: Postscript, PDF, BibTeX

Publications