Refine
Document Type
- Conference proceeding (18)
- Journal article (3)
- Book chapter (3)
- Doctoral Thesis (1)
Language
- English (25)
Has full text
- yes (25)
Is part of the Bibliography
- yes (25)
Institute
- Informatik (25)
Publisher
Massive data transfers in modern key/value stores resulting from low data-locality and data-to-code system design hurt their performance and scalability. Near-data processing (NDP) designs represent a feasible solution, which although not new, have yet to see widespread use.
In this paper we introduce nKV, which is a key/value store utilizing native computational storage and near-data processing. On the one hand, nKV can directly control the data and computation placement on the underlying storage hardware. On the other hand, nKV propagates the data formats and layouts to the storage device where, software and hardware parsers and accessors are implemented. Both allow NDP operations to execute in host-intervention-free manner, directly on physical addresses and thus better utilize the underlying hardware. Our performance evaluation is based on executing traditional KV operations (GET, SCAN) and on complex graph-processing algorithms (Betweenness Centrality) in-situ, with 1.4×-2.7× better performance on real hardware – the COSMOS+ platform.
Over the last decades, a tremendous change toward using information technology in almost every daily routine of our lives can be perceived in our society, entailing an incredible growth of data collected day-by-day on Web, IoT, and AI applications.
At the same time, magneto-mechanical HDDs are being replaced by semiconductor storage such as SSDs, equipped with modern Non-Volatile Memories, like Flash, which yield significantly faster access latencies and higher levels of parallelism. Likewise, the execution speed of processing units increased considerably as nowadays server architectures comprise up to multiple hundreds of independently working CPU cores along with a variety of specialized computing co-processors such as GPUs or FPGAs.
However, the burden of moving the continuously growing data to the best fitting processing unit is inherently linked to today’s computer architecture that is based on the data-to-code paradigm. In the light of Amdahl's Law, this leads to the conclusion that even with today's powerful processing units, the speedup of systems is limited since the fraction of parallel work is largely I/O-bound.
Therefore, throughout this cumulative dissertation, we investigate the paradigm shift toward code-to-data, formally known as Near-Data Processing (NDP), which relieves the contention on the I/O bus by offloading processing to intelligent computational storage devices, where the data is originally located.
Firstly, we identified Native Storage Management as the essential foundation for NDP due to its direct control of physical storage management within the database. Upon this, the interface is extended to propagate address mapping information and to invoke NDP functionality on the storage device. As the former can become very large, we introduce Physical Page Pointers as one novel NDP abstraction for self-contained immutable database objects.
Secondly, the on-device navigation and interpretation of data are elaborated. Therefore, we introduce cross-layer Parsers and Accessors as another NDP abstraction that can be executed on the heterogeneous processing capabilities of modern computational storage devices. Thereby, the compute placement and resource configuration per NDP request is identified as a major performance criteria. Our experimental evaluation shows an improvement in the execution durations of 1.4x to 2.7x compared to traditional systems. Moreover, we propose a framework for the automatic generation of Parsers and Accessors on FPGAs to ease their application in NDP.
Thirdly, we investigate the interplay of NDP and modern workload characteristics like HTAP. Therefore, we present different offloading models and focus on an intervention-free execution. By propagating the Shared State with the latest modifications of the database to the computational storage device, it is able to process data with transactional guarantees. Thus, we achieve to extend the design space of HTAP with NDP by providing a solution that optimizes for performance isolation, data freshness, and the reduction of data transfers. In contrast to traditional systems, we experience no significant drop in performance when an OLAP query is invoked but a steady and 30% faster throughput.
Lastly, in-situ result-set management and consumption as well as NDP pipelines are proposed to achieve flexibility in processing data on heterogeneous hardware. As those produce final and intermediary results, we continue investigating their management and identified that an on-device materialization comes at a low cost but enables novel consumption modes and reuse semantics. Thereby, we achieve significant performance improvements of up to 400x by reusing once materialized results multiple times.
Blockchains yield to new workloads in database management systems and K/V-stores. Distributed Ledger Technology (DLT) is a technique for managing transactions in ’trustless’ distributed systems. Yet, clients of nodes in blockchain networks are backed by ’trustworthy’ K/V-Stores, like LevelDB or RocksDB in Ethereum, which are based on Log-Structured Merge Trees (LSM Trees). However, LSM-Trees do not fully match the properties of blockchains and enterprise workloads.
In this paper, we claim that Partitioned B-Trees (PBT) fit the properties of this DLT: uniformly distributed hash keys, immutability, consensus, invalid blocks, unspent and off-chain transactions, reorganization and data state / version ordering in a distributed log-structure. PBT can locate records of newly inserted key-value pairs, as well as data of unspent transactions, in separate partitions in main memory. Once several blocks acquire consensus, PBTs evict a whole partition, which becomes immutable, to secondary storage. This behavior minimizes write amplification and enables a beneficial sequential write pattern on modern hardware. Furthermore, DLT implicate some type of log-based versioning. PBTs can serve as MV-store for data storage of logical blocks and indexing in multi-version concurrency control (MVCC) transaction processing.
Characteristics of modern computing and storage technologies fundamentally differ from traditional hardware. There is a need to optimally leverage their performance, endurance and energy consumption characteristics. Therefore, existing architectures and algorithms in modern high performance database management systems have to be redesigned and advanced. Multi Version Concurrency Control (MVCC) approaches in data-base management systems maintain multiple physically independent tuple versions. Snapshot isolation approaches enable high parallelism and concurrency in workloads with almost serializable consistency level. Modern hardware technologies benefit from multi-version approaches. Indexing multi-version data on modern hardware is still an open research area. In this paper, we provide a survey of popular multi-version indexing approaches and an extended scope of high performance single-version approaches. An optimal multi-version index structure brings look-up efficiency of tuple versions, which are visible to transactions, and effort on index maintenance in balance for different workloads on modern hardware technologies.
Database management systems (DBMS) are critical performance components in large scale applications under modern update intensive workloads. Additional access paths accelerate look-up performance in DBMS for frequently queried attributes, but the required maintenance slows down update performance. The ubiquitous B+ tree is a commonly used key-indexed access path that is able to support many required functionalities with logarithmic access time to requested records. Modern processing and storage technologies and their characteristics require reconsideration of matured indexing approaches for today's workloads. Partitioned B-trees (PBT) leverage characteristics of modern hardware technologies and complex memory hierarchies as well as high update rates and changes in workloads by maintaining partitions within one single B+-Tree. This paper includes an experimental evaluation of PBTs optimized write pattern and performance improvements. With PBT transactional throughput under TPC-C increases 30%; PBT results in beneficial sequential write patterns even in presence of updates and maintenance operations.
Database Management Systems (DBMS) need to handle large updatable datasets in on-line transaction processing (OLTP) workloads. Most modern DBMS provide snapshots of data in multi-version concurrency control (MVCC) transaction management scheme. Each transaction operates on a snapshot of the database, which is calculated from a set of tuple versions. High parallelism and resource-efficient append-only data placement on secondary storage is enabled. One major issue in indexing tuple versions on modern hardware technologies is the high write amplification for tree-indexes.
Partitioned B-Trees (PBT) [5] is based on the structure of the ubiquitous B+ Tree [8]. They achieve a near optimal write amplification and beneficial sequential writes on secondary storage. Yet they have not been implemented in a MVCC enabled DBMS to date.
In this paper we present the implementation of PBTs in PostgreSQL extended with SIAS. Compared to PostgreSQL’s B+–Trees PBTs have 50% better transaction throughput under TPC-C and a 30% improvement to standard PostgreSQL with Heap-Only Tuples.
Modern mixed (HTAP)workloads execute fast update-transactions and long running analytical queries on the same dataset and system. In multi-version (MVCC) systems, such workloads result in many short-lived versions and long version-chains as well as in increased and frequent maintenance overhead.
Consequently, the index pressure increases significantly. Firstly, the frequent modifications cause frequent creation of new versions, yielding a surge in index maintenance overhead. Secondly and more importantly, index-scans incur extra I/O overhead to determine, which of the resulting tuple versions are visible to the executing transaction (visibility-check) as current designs only store version/timestamp information in the base table – not in the index. Such index-only visibility-check is critical for HTAP workloads on large datasets.
In this paper we propose the Multi Version Partitioned B-Tree (MV-PBT) as a version-aware index structure, supporting index-only visibility checks and flash-friendly I/O patterns. The experimental evaluation indicates a 2x improvement for analytical queries and 15% higher transactional throughput under HTAP workloads. MV-PBT offers 40% higher tx. throughput compared to WiredTiger’s LSM-Tree implementation under YCSB.
Active storage
(2018)
In brief, Active Storage refers to an architectural hardware and software paradigm, based on collocation storage and compute units. Ideally, it will allow to execute application-defined data ... within the physical data storage. Thus Active Storage seeks to minimize expensive data movement, improving performance, scalability, and resource efficiency. The effective use of Active Storage mandates new architectures, algorithms, interfaces, and development toolchains.
A transaction is a demarcated sequence of application operations, for which the following properties are guaranteed by the underlying transaction processing system (TPS): atomicity, consistency, isolation, and durability (ACID). Transactions are therefore a general abstraction, provided by TPS that simplifies application development by relieving transactional applications from the burden of concurrency and failure handling. Apart from the ACID properties, a TPS must guarantee high and robust performance (high transactional throughput and low response times), high reliability (no data loss, ability to recover last consistent state, fault tolerance), and high availability (infrequent outages, short recovery times).
The architectures and workhorse algorithms of a high-performance TPS are built around the properties of the underlying hardware. The introduction of nonvolatile memories (NVM) as novel storage technology opens an entire new problem space, with the need to revise aspects such as the virtual memory hierarchy, storage management and data placement, access paths, and indexing. NVM are also referred to as storage-class memory (SCM).
In the present tutorial we perform a cross-cut analysis of database storage management from the perspective of modern storage technologies. We argue that neither the design of modern DBMS, nor the architecture of modern storage technologies are aligned with each other. Moreover, the majority of the systems rely on a complex multi-layer and compatibility oriented storage stack. The result is needlessly suboptimal DBMS performance, inefficient utilization, or significant write amplification due to outdated abstractions and interfaces. In the present tutorial we focus on the concept of native storage, which is storage operated without intermediate abstraction layers over an open native storage interface and is directly controlled by the DBMS.