No abstract

Deduplication reduces the size of the data stored in large-scale storage systems by replacing duplicate data blocks with references to their unique copies. This creates dependencies between files that contain similar content, and complicates the management of data in the system. In this paper, we address the problem of data migration, where files are remapped between different volumes as a result of system expansion or maintenance. The challenge of determining which files and blocks to migrate has been studied extensively for systems without deduplication. In the context of deduplicated storage, however, only simplified migration scenarios have been considered. In this paper, we formulate the general migration problem for deduplicated systems as an optimization problem whose objective is to minimize the system’s size while ensuring that the storage load is evenly distributed between the system’s volumes, and that the network traffic required for the migration does not exceed its allocation. We then present three algorithms for generating effective migration plans, each based on a different approach and representing a different tradeoff between computation time and migration efficiency. Our greedy algorithm provides modest space savings, but is appealing thanks to its exceptionally short runtime. Its results can be improved by using larger system representations. Our theoretically optimal algorithm formulates the migration problem as an ILP (integer linear programming) instance. Its migration plans consistently result in smaller and more balanced systems than those of the greedy approach, although its runtime is long and, as a result, the theoretical optimum is not always found. Our clustering algorithm enjoys the best of both worlds: its migration plans are comparable to those generated by the ILP-based algorithm, but its runtime is shorter, sometimes by an order of magnitude. It can be further accelerated at a modest cost in the quality of its results.

Abstract. Although sophisticated runtime bug detection tools exist to root out several kinds of concurrency errors, they cannot easily be used at the kernel level. Our Redflag framework and system seeks to bring these essential techniques to the Linux kernel by addressing issues faced by other tools. First, other tools typically examine every potentially concurrent memory access, which is infeasible in the kernel because of the overhead it would introduce. Redflag minimizes overhead by using offline analysis together with an efficient in-line logging system and by supporting targeted configurable logging of specific kernel components and data structures. Targeted analysis reduces overhead and avoids presenting developers with error reports for components they are not responsible for. Second, other tools do not take into account some of the synchronization patterns found in the kernel, resulting in false positives. We explore two algorithms for detecting concurrency errors: one for race conditions and another for atomicity violations; we enhanced them to take into account some specifics of synchronization in the kernel. In particular, we introduce Lexical Object Availability (LOA) analysis to deal with multi-stage escape and other complex order-enforcing synchronization. We evaluate the effectiveness and performance of Redflag on two file systems and a video driver.

No abstract

No abstract