Refine
Year of publication
Document Type
- Other (879) (remove)
Language
- English (643)
- German (221)
- Spanish (5)
- Italian (4)
- Multiple languages (2)
- Polish (2)
- French (1)
- Portuguese (1)
Keywords
- Arrayseismologie (5)
- array seismology (5)
- Dysphagie (4)
- E-Learning (4)
- Erdbeben (4)
- Judaism (4)
- Judentum (4)
- MOOC (4)
- Patholinguistik (4)
- Schluckstörung (4)
Institute
- Institut für Biochemie und Biologie (86)
- Hasso-Plattner-Institut für Digital Engineering GmbH (83)
- Institut für Physik und Astronomie (82)
- Institut für Geowissenschaften (73)
- Institut für Mathematik (46)
- Department Psychologie (44)
- Department Sport- und Gesundheitswissenschaften (44)
- Institut für Ernährungswissenschaft (31)
- Hasso-Plattner-Institut für Digital Engineering gGmbH (30)
- Institut für Chemie (29)
Ästhetische Intelligenz
(1998)
Zenralamerika
(2011)
Modern server systems with large NUMA architectures necessitate (i) data being distributed over the available computing nodes and (ii) NUMA-aware query processing to enable effective parallel processing in database systems. As these architectures incur significant latency and throughout penalties for accessing non-local data, queries should be executed as close as possible to the data. To further increase both performance and efficiency, data that is not relevant for the query result should be skipped as early as possible. One way to achieve this goal is horizontal partitioning to improve static partition pruning. As part of our ongoing work on workload-driven partitioning, we have implemented a recent approach called aggressive data skipping and extended it to handle both analytical as well as transactional access patterns. In this paper, we evaluate this approach with the workload and data of a production enterprise system of a Global 2000 company. The results show that over 80% of all tuples can be skipped in average while the resulting partitioning schemata are surprisingly stable over time.
Workload-Driven Fragment Allocation for Partially Replicated Databases Using Linear Programming
(2019)
In replication schemes, replica nodes can process read-only queries on snapshots of the master node without violating transactional consistency. By analyzing the workload, we can identify query access patterns and replicate data depending to its access frequency. In this paper, we define a linear programming (LP) model to calculate the set of partial replicas with the lowest overall memory capacity while evenly balancing the query load. Furthermore, we propose a scalable decomposition heuristic to calculate solutions for larger problem sizes. While guaranteeing the same performance as state-of-the-art heuristics, our decomposition approach calculates allocations with up to 23% lower memory footprint for the TPC-H benchmark.
Words as social tools
(2019)