Peter Bumbulis

SOFORT: a hybrid SCM-DRAM storage engine for fast data recovery

Storage Class Memory (SCM) has the potential to significantly improve database performance. This potential has been well documented for throughput [4] and response time [25, 22]. In this paper we show that SCM has also the potential to significantly improve restart performance, a shortcoming of traditional main memory database systems. We present SOFORT, a hybrid SCM-DRAM storage engine that leverages full capabilities of SCM by doing away with a traditional log and updating the persisted data in place in small increments. We show that we can achieve restart times of a few seconds independent of instance size and transaction volume without significantly impacting transaction throughput.

SQL Anywhere: A Holistic Approach to Database Self-management

In this paper we present an overview of the self-management features of SQL anywhere, a full-function relational database system designed for frontline business environments with minimal administration. SQL Anywhere can serve as a high-performance workgroup server, an embedded database that is installed along with an application, or as a mobile database installed on a handheld device that provides full database services, including two-way synchronization, to applications when the device is disconnected from the corporate intranet. We illustrate how the various self-management features work in concert to provide a robust data management solution in zero-administration environments.

Towards scalable real-time analytics: an architecture for scale-out of OLxP workloads

We present an overview of our work on the SAP HANA Scale-out Extension, a novel distributed database architecture designed to support large scale analytics over real-time data. This platform permits high performance OLAP with massive scale-out capabilities, while concurrently allowing OLTP workloads. This dual capability enables analytics over real-time changing data and allows fine grained user-specified service level agreements (SLAs) on data freshness. We advocate the decoupling of core database components such as query processing, concurrency control, and persistence, a design choice made possible by advances in high-throughput low-latency networks and storage devices. We provide full ACID guarantees and build on a logical timestamp mechanism to provide MVCC-based snapshot isolation, while not requiring synchronous updates of replicas. Instead, we use asynchronous update propagation guaranteeing consistency with timestamp validation. We provide a view into the design and development of a large scale data management platform for real-time analytics, driven by the needs of modern enterprise customers.

Automatic tuning of the multiprogramming level in Sybase SQL Anywhere

This paper looks at the problem of automatically tuning the database server multiprogramming level to improve database server performance under varying workloads. We describe two tuning algorithms that were considered and how they performed under different workloads. We then present the hybrid approach that we have successfully implemented in SQL Anywhere 12. We found that the hybrid approach yielded better performance than each of the algorithms separately.

A compact B-tree

In this paper we describe a Patricia tree-based B-tree variant suitable for OLTP. In this variant, each page of the B-tree contains a local Patricia tree instead of the usual sorted array of keys. It has been implemented in iAnywhere ASA Version 8.0. Preliminary experience has shown that these indexes can provide significant space and performance benefits over existing ASA indexes.

Page As You Go: Piecewise Columnar Access In SAP HANA

In-memory columnar databases such as SAP HANA achieve extreme performance by means of vector processing over logical units of main memory resident columns. The core in-memory algorithms can be challenged when the working set of an application does not fit into main memory. To deal with memory pressure, most in-memory columnar databases evict candidate columns (or tables) using a set of heuristics gleaned from recent workload. As an alternative approach, we propose to reduce the unit of load and eviction from column to a contiguous portion of the in-memory columnar representation, which we call a page. In this paper, we adapt the core algorithms to be able to operate with partially loaded columns while preserving the performance benefits of vector processing. Our approach has two key advantages. First, partial column loading reduces the mandatory memory footprint for each column, making more memory available for other purposes. Second, partial eviction extends the in-memory lifetime of partially loaded column. We present a new in-memory columnar implementation for our approach, that we term page loadable column. We design a new persistency layout and access algorithms for the encoded data vector of the column, the order-preserving dictionary, and the inverted index. We compare the performance attributes of page loadable columns with those of regular in-memory columns and present a use-case for page loadable columns for cold data in data aging scenarios. Page loadable columns are completely integrated in SAP HANA, and we present extensive experimental results that quantify the performance overhead and the resource consumption when these columns are deployed.

Efficient and consistent replication for distributed logs

Distributed shared logs are a powerful building block for distributed systems. By providing fault-tolerant persistence and strong ordering guarantees, applications can use a distributed shared log to reliably communicate a stream of events between processes. This can be used, for example, to replicate application state or to build a reliable publish/subscribe system. The log itself must also replicate data in order to provide availability and fault-tolerance. Key to the design of a distributed shared log is the choice of replication algorithm, which will determine many properties of the system. We propose an algorithm for consistent replication of log data, quorum-replication with meta-data exchange (QMX), that is linearizable while allowing writes to be successful with only a single round-trip to a quorum of replicas and allowing reads to generally be serviced by any single replica, or read-one/write-quorum. This is achieved by coupling the reads with an asynchronous message exchange algorithm that continuously runs amongst the replicas. The message exchange algorithm allows replicas to infer the global state of writes across the cluster, in order to deduce which writes have been successfully quorum replicated and which have not. This metadata allows any single replica to directly answer reads in many cases, though in the worst case a read must wait for the message passing round to complete before being serviced which requires a majority quorum of servers to be responsive.