A Query Processing Mechanism Based on the Broadcast Queue for Broadcast Database Systems

2006 ◽  
Author(s):  
S. Kitajima ◽  
Jing Cai ◽  
T. Terada ◽  
T. Hara ◽  
S. Nishio
Keyword(s):  
Query Processing ◽  
Database Systems ◽  
Processing Mechanism

scholarly journals Smart Intra-query Fault Tolerance for Massive Parallel Processing Databases

2019 ◽  
Vol 5 (1) ◽  
pp. 65-79
Author(s):  
Yunhong Ji ◽  
Yunpeng Chai ◽  
Xuan Zhou ◽  
Lipeng Ren ◽  
Yajie Qin
Keyword(s):  
Cost Effectiveness ◽  
Fault Tolerance ◽  
Parallel Processing ◽  
Query Processing ◽  
Success Rate ◽  
Database Systems ◽  
Analytical Processing ◽  
Commodity Clusters ◽  
Massive Parallel Processing ◽  
Query Latency

AbstractIntra-query fault tolerance has increasingly been a concern for online analytical processing, as more and more enterprises migrate data analytical systems from mainframes to commodity computers. Most massive parallel processing (MPP) databases do not support intra-query fault tolerance. They may suffer from prolonged query latency when running on unreliable commodity clusters. While SQL-on-Hadoop systems can utilize the fault tolerance support of low-level frameworks, such as MapReduce and Spark, their cost-effectiveness is not always acceptable. In this paper, we propose a smart intra-query fault tolerance (SIFT) mechanism for MPP databases. SIFT achieves fault tolerance by performing checkpointing, i.e., materializing intermediate results of selected operators. Different from existing approaches, SIFT aims at promoting query success rate within a given time. To achieve its goal, it needs to: (1) minimize query rerunning time after encountering failures and (2) introduce as less checkpointing overhead as possible. To evaluate SIFT in real-world MPP database systems, we implemented it in Greenplum. The experimental results indicate that it can improve success rate of query processing effectively, especially when working with unreliable hardware.

Heterogeneous placement optimization for database query processing

2017 ◽  
Vol 59 (3) ◽  
Author(s):  
Tomas Karnagel ◽  
Dirk Habich
Keyword(s):  
Query Processing ◽  
Database Systems ◽  
Improve Performance ◽  
Database Query ◽  
Placement Decisions ◽  
Placement Optimization ◽  
Database Query Processing ◽  
The Given

AbstractComputing hardware is constantly evolving and database systems need to adapt to ongoing hardware changes to improve performance. The current hardware trend is heterogeneity, where multiple computing units like CPUs and GPUs are used together in one system. In this paper, we summarize our efforts to use hardware heterogeneity efficiently for query processing. We discuss different approaches of execution and investigate heterogeneous placement in detail by showing, how to automatically determine operator placement decisions according to the given hardware environment and query properties.

A TWO-LEVEL METADATA DICTIONARY APPROACH FOR SEMANTIC QUERY PROCESSING IN MULTIDATABASE SYSTEMS

1993 ◽  
Vol 03 (02) ◽  
pp. 231-255
Author(s):  
MIIN-JENG PAN ◽  
SHI-KUO CHANG ◽  
CHIEN-CHIAO YANG
Keyword(s):  
Query Processing ◽  
Database Systems ◽  
Database Integration ◽  
Inference Control ◽  
Semantic Query ◽  
Multidatabase System ◽  
Integration Strategy ◽  
Experimental Implementation ◽  
Schema Translation ◽  
Two Phases

A multidatabase system (MDBS) is a system that integrates several autonomous database systems and provides users with a uniform access to all the databases. In this paper we developed a two-level active metadata dictionary approach for semantic query processing. To capture the global view of data schemas of participating databases which may be heterogeneous, a Hornclause data model is used. The lower-level metadata dictionaries (LLMDs) keep metadata for each corresponding local database in MDBS. The higher-level metadata dictionary (HLMD) integrates the metadata about all LLMDs. The database integration strategy includes two phases: schema translation and schema integration. It is a bottom-up approach integrating schema from the underlying database schemas. The evaluation strategy is a top-down approach. It starts with a query as a global goal to be achieved, unifies and optimizes the query to decompose the goal into subgoals that can be evaluated against extensional database, then translates these subgoals into corresponding queries against underlying DBMSs. To solve the control problem, we employ a G-net model for procedure control and inference control. An experimental implementation in Prolog is described.

Sign in / Sign up

Export Citation Format

Share Document