Research output: Contribution to journal › Conference article › peer-review
Big Data Virtualization: Why and How? / Bogdanov, Alexander ; Degtyarev, Alexander ; Shchegoleva, Nadezhda ; Korkhov, Vladimir ; Khvatov, Valery .
In: CEUR Workshop Proceedings, Vol. 2679, 2020, p. 11-21.Research output: Contribution to journal › Conference article › peer-review
}
TY - JOUR
T1 - Big Data Virtualization: Why and How?
AU - Bogdanov, Alexander
AU - Degtyarev, Alexander
AU - Shchegoleva, Nadezhda
AU - Korkhov, Vladimir
AU - Khvatov, Valery
PY - 2020
Y1 - 2020
N2 - The increasing variability, dynamic and heterogenous nature of big data, as well as the flexibility required by data producers and consumers lead to the necessity of organizing access to data without requiring information about its structure or belonging to any particular information system, i.e. data virtualization. Data virtualization complements the concept of a virtual supercomputer, allowing us to consider the computing environment as a single information continuum, where computing tools are part of the general data model and data is considered in the context of three basic components: Integration, analysis, and data presentation. In this paper, we present the concept of unified, generalized and encapsulated representation of data coming from a heterogenous set of data sources, based on the extension of the Logical Data Warehouse (LDW) and the Distributed Data Network in the form of a distributed ledger a- the virtual DLT (vDLT). The main difference between the vDLT and LDW approaches is the decentralized management of data using a consensus mechanism. We discuss data virtualization practices, the methodology of constructing a virtualized data environment, and compare core steps and approaches for each of the two directions.
AB - The increasing variability, dynamic and heterogenous nature of big data, as well as the flexibility required by data producers and consumers lead to the necessity of organizing access to data without requiring information about its structure or belonging to any particular information system, i.e. data virtualization. Data virtualization complements the concept of a virtual supercomputer, allowing us to consider the computing environment as a single information continuum, where computing tools are part of the general data model and data is considered in the context of three basic components: Integration, analysis, and data presentation. In this paper, we present the concept of unified, generalized and encapsulated representation of data coming from a heterogenous set of data sources, based on the extension of the Logical Data Warehouse (LDW) and the Distributed Data Network in the form of a distributed ledger a- the virtual DLT (vDLT). The main difference between the vDLT and LDW approaches is the decentralized management of data using a consensus mechanism. We discuss data virtualization practices, the methodology of constructing a virtualized data environment, and compare core steps and approaches for each of the two directions.
KW - Big data
KW - Data virtualization
KW - Virtual Personal Supercomputer
KW - data network
KW - data marketplaces
KW - Distributed ledger technologies
KW - Big Data
KW - distributed ledger technologies
UR - http://www.scopus.com/inward/record.url?scp=85092302928&partnerID=8YFLogxK
M3 - Conference article
VL - 2679
SP - 11
EP - 21
JO - CEUR Workshop Proceedings
JF - CEUR Workshop Proceedings
SN - 1613-0073
Y2 - 8 June 2020 through 10 June 2020
ER -
ID: 62372498