Relational query processors derive much of their effectiveness from the awareness of specific table properties like sort order, size, or absence of duplicate tuples. This text applies (and adapts) this successful principle to database-supported XML and XPath processing: the relational system is made tree aware, i.e., tree properties like subtree size, intersection of paths, inclusion or disjointness of subtrees are made explicit. We propose a local change to the database kernel, the staircase join, which encapsulates the necessary tree knowledge needed to improve XPath performance. Staircase join

This paper introduces the FAWN—Fast Array of Wimpy Nodes—cluster architecture for providing fast, scalable, and power-efficient key-value storage. A FAWN links together a large number of tiny nodes built using embedded processors and small amounts (2–16GB) of flash memory into an ensemble capable of handling 700 queries per second per node, while consuming fewer than 6 watts of power per node. We have designed and implemented a clustered key-value storage system, FAWN-DHT, that runs atop these node. Nodes in FAWN-DHT use a specialized log-like back-end hash-based database to ensure that the system can absorb the large write workload imposed by frequent node arrivals and departures. FAWN uses a two-level cache hierarchy to ensure that imbalanced workloads cannot create hot-spots on one or a few wimpy nodes that impair the system’s ability to service queries at its guaranteed rate. Our evaluation of a small-scale FAWN cluster and several candidate FAWN node systems suggest that FAWN can be a practical approach to building large-scale storage for seek-intensive workloads. Our further analysis indicates that a FAWN cluster is cost-competitive with other approaches (e.g., DRAM, multitudes of magnetic disks, solid-state disk) to providing high query rates, while consuming 3-10x less power. Acknowledgements: We thank the members and companies of the CyLab Corporate Partners and the PDL

Solid state drives perform random reads more than 100x faster than traditional magnetic hard disks, while offering comparable sequential read and write bandwidth. Because of their potential to speed up applications, as well as their reduced power consumption, these new drives are expected to gradually replace hard disks as the primary permanent storage media in large data centers. However, although they may benefit applications that stress random reads immediately, they may not improve database applications, especially those running long data analysis queries. Database query processing engines have been designed around the speed mismatch between random and sequential I/O on hard disks and their algorithms currently emphasize sequential accesses for disk-resident data. In this paper, we investigate data structures and algorithms that leverage fast random reads to speed up selection, projection, and join operations in relational query processing. We first demonstrate how a column-based layout within each page reduces the amount of data read during selections and projections. We then introduce FlashJoin, a general pipelined join algorithm that minimizes accesses to base and intermediate relational data. FlashJoin’s binary join kernel accesses only the join attributes, producing partial results in the form of a join index. Subsequently, its fetch kernel retrieves the attributes for later nodes in the query plan as they are needed. FlashJoin significantly reduces memory and I/O requirements for each join in the query. We implemented these techniques inside Postgres and experimented with an enterprise SSD drive. Our techniques improved query runtimes by up to 6x for queries ranging from simple relational scans and joins to full TPC-H queries.

Secure chips, e.g. present in smart cards, TPM, USB dongles are now ubiquitous in applications with strong security requirements. Secure chips host personal data that must be carefully managed and protected, thus requiring embedded data management techniques. However, secure chips have severe hardware constraints which make traditional database techniques irrelevant. We previously addressed the problem of scaling down database techniques for the smart card and proposed the design of a DBMS kernel called PicoDBMS. This paper summarizes the learning obtained during an extensive performance analysis of PicoDBMS. Then it studies to which extent secure chips hardware evolution should impact the design of embedded data management techniques. Finally, it draws research perspectives linked to a broader usage of secure chip data management techniques. 1

We present algorithms using graphics processing units (GPUs) to efficiently perform database management queries. Our algorithms use efficient data memory representations and storage models on GPUs to perform fast database computations. We present relational database algorithms that successfully exploit the high memory bandwidth and the inherent parallelism available in GPUs. We implement these algorithms on commodity GPUs and compare their performance with optimized CPU-based algorithms. We show that the GPUs can be used as a co-processor to accelerate many database and data mining queries. 1.

An increasing amount of personal data is automatically gathered and stored on servers by administrations, hospitals, insurance companies, etc. Citizen themselves often count on internet companies to store their data and make them reliable and highly available through the internet. However, these benefits must be weighed against privacy risks incurred by centralization. This paper suggests a radically different way of considering the management of personal data. It builds upon the emergence of new portable and secure devices combining the security of smart cards and the storage capacity of NAND Flash chips. By embedding a full-fledged Personal Data Server in such devices, user control of how her sensitive data is shared by others (by whom, for how long, according to which rule, for which purpose) can be fully reestablished and convincingly enforced. To give sense to this vision, Personal Data Servers must be able to interoperate with external servers and must provide traditional database services like durability, availability, query facilities, transactions. This paper proposes an initial design for the Personal Data Server approach, identifies the main technical challenges associated with it and sketches preliminary solutions. We expect that this paper will open exciting perspectives for future database research. 1.

Table scans have become more interesting recently due to greater use of ad-hoc queries and greater availability of multicore, vector-enabled hardware. Table scan performance is limited by value representation, table layout, and processing techniques. In this paper we propose a new layout and processing technique for efficient one-pass predicate evaluation. Starting with a set of rows with a fixed number of bits per column, we append columns to form a set of banks and then pad each bank to a supported machine word length, typically 16, 32, or 64 bits. We then evaluate partial predicates on the columns of each bank, using a novel evaluation strategy that evaluates column level equality, range tests, IN-list predicates, and conjuncts of these predicates, simultaneously on multiple columns within a bank, and on multiple rows within a machine register. This approach outperforms pure column stores, which must evaluate the partial predicates one column at a time. We evaluate and compare the performance and representation overhead of this new approach and several proposed alternatives. 1.

Database management systems (DBMSes) form a cornerstone of modern IT infrastructure, and it is essential that they have excellent performance. Much of the work to date on optimizing DBMS performance has emphasized ensuring efficient data access from secondary storage. This paper shows that DBMSes can also benefit significantly from dynamic code specialization. Our approach focuses on the iterative query evaluation loops typically used by such systems. Query evaluation involves extensive references to the relational schema, predicate values, and join types, which are all invariant during query evaluation, and thus are subject to dynamic value-based code specialization. We introduce three distinct types of specialization, each corresponding to a particular kind of invariant. We realize these techniques, in concert termed micro-specialization, via a DBMS-independent run-time environment and apply them to a highperformance open-source DBMS, PostgreSQL. We show that microspecialization requires minimal changes to the DBMS and can yield performance improvements simultaneously across a wide range of queries and modifications, in terms of storage, CPU usage, and I/O time of standard DBMS benchmarks. We also discuss an integrated development environment that helps DBMS developers apply micro-specializations to identified target code sequences. 1.

OLTP and transactional workloads are increasingly common in computer systems, ranging from ecommerce to warehousing to inventory management. It is valuable to provide priority scheduling in these systems, to reduce the response time for the most important clients, e.g. the "big spenders". Two-phase locking, commonly used in DBMS, makes prioritization di#cult, as transactions wait for locks held by others regardless of priority. Common lock scheduling solutions, including non-preemptive priority inheritance and preemptive abort, have performance drawbacks for TPC-C type workloads.

Flash memory affects not only storage options but also query processing. In this paper, we analyze the use of flash memory for database query processing, including algorithms that combine flash memory and traditional disk drives. We first focus on flash-resident databases and present data structures and algorithms that leverage the fast random reads of flash to speed up selection, projection, and join operations. FlashScan and FlashJoin are two such algorithms that leverage a column-based layout to significantly reduce memory and I/O requirements. Experiments with Postgres and an enterprise SSD drive show improved query runtimes by up to 6x for queries ranging from simple relational scans and joins to full TPC-H queries. In the second part of the paper, we use external merge sort as a prototypical query execution algorithm to demonstrate that the most advantageous external sort algorithms combine flash memory and traditional disk, exploiting the fast access latency of flash memory as well as the fast transfer bandwidth and inexpensive capacity of traditional disks. Looking forward, database query processing in a three-level memory hierarchy of RAM, flash memory, and traditional disk can be generalized to any number of levels that future hardware may feature. 1