English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Multidimensional data are becoming more prevalent, partly due to the rise ofthe Internet of Things (IoT), and with that the need to ingest and analyze datastreams at rates higher than before. Some industrial IoT applications requireingesting millions of records per second, while processing queries on recentlyingested and historical data. Unfortunately, existing database systems suitedto multidimensional data exhibit low per-node ingestion performance, and evenif they can scale horizontally in distributed settings, they require largenumber of nodes to meet such ingest demands. For this reason, in this paper weevaluate a single-node multidimensional data store for high-velocity sensordata. Its design centers around a two-level indexing structure, wherein theglobal index is an in-memory R*-tree and the local indices are serializedkd-trees. This study is confined to records with numerical indexing fields andrange queries, and covers ingest throughput, query response time, and storagefootprint. We show that the adopted design streamlines data ingestion andoffers ingress rates two orders of magnitude higher than those of PerconaServer, SQLite, and Druid. Our prototype also reports query response timescomparable to or better than those of Percona Server and Druid, and comparesfavorably in terms of storage footprint. In addition, we evaluate a kd-treepartitioning based scheme for grouping incoming streamed data records. Comparedto a random scheme, this scheme produces less overlap between groups ofstreamed records, but contrary to what we expected, such reduced overlap doesnot translate into better query performance. By contrast, the local indicesprove much more beneficial to query performance. We believe the experiencereported in this paper is valuable to practitioners and researchers alikeinterested in building database systems for high-velocity multidimensionaldata.

Ingestion, Indexing and Retrieval of High-Velocity Multidimensional Sensor Data on a Single Node