Forrester Research Report: Information fabric--Enterprise Data Virtualization (part VIII, digest translation)

The centrally Managed distributed Metadata Repository (DMR)
Centralized-managed distributed metadata Storage (DMR)

The greatest challenge with distributed data repositories are integration, especially when dealing with large amounts of Di Sparate data and integrating them. The biggest challenge of distributed data storage is how to integrate, especially real-time integration, and handle large amounts of heterogeneous data. Unlike EII, ETL, and data replication approaches to data integration in which all data access has to is predefined with a DMR, distributed data is integrated at runtime and requires no predefined access paths, thus providing greater flexibilit Y (Figure 3). Unlike EII, ETL, and data replication methods, each data access to data integration requires a predefined DMR, and the runtime's distributed data integration does not require a predefined access path, and is more flexible (see figure below). Unfortunately, the industry still lacks comprehensive standards or products to tie various metadata. Unfortunately, the industry currently lacks a complete set of standards or products to combine different meta data.

How a centrally managed DMR different than EII? What is the difference between DMR and EII in centralized management? EII is a broad term this covers a collection of technologies and best practices for providing custom views into multiple D ATA sources to integrate data and content for real-time read and write access by applications. EII is a broader concept that encompasses the technology sets and best practices that are used to derive the majority of applications for real-time read-write access from defining views. EII integrates data via views, fetching data in real time as needed instead of requiring a involved data movement process Like ETL. EII through the view integration of data, on-demand real-time access to data, without the same as the ETL data handling process. Unlike information fabric, EII is mostly focused on structured data via SQL relational access and therefore usually can ' t Include content. Unlike information fabric, the EII is primarily focused on relational access to structured data through SQL, so it usually does not contain content. eii

The EII approach to integration can is a slow process, because it often requires EII platforms to communicate with MULTIPL E source systems, fetch multiple result sets, and merge them to a single result that application receives. Integration with EII can be a slow process because it typically requires the EII platform to communicate with multiple source systems, query multiple result sets, and consolidate them into application-accessible uniform results. With information fabric, the data is represented by virtual metadata constructs, and data access can be constructed and altere D in real time to cope with evolving business requirements. With information Fabric, data is represented through a virtual metadata structure, and access to the data can be modified and built in real time to meet rising business needs. The DMR component integrates and caches data to optimize data access. The DMR component integrates data into real time and uses data caching to optimize data access.

The key characteristics of a DMR include:
The main features of DMR are:

Flexible repository architecture. Flexible storage architecture. The DMR should is able to adapt to new applications, and changing business requirements-meaning, it metamodel shoul D is extensible, should integrate with the sources of metadata where possible, and should provide tooling as needed to C Onfigure the fabric. DMR needs to adapt to new applications and changing business requirements-which means that the metadata model must be scalable to integrate with other possible metadata sources, as well as provide tools for configuring fabric.

distribution across servers. The distribution across servers. Metadata must is stored on all servers participating in the information fabric and synchronized in real time across server S. The metadata needs to be stored in the individual servers of the information fabric, and can be synchronized across servers in real time.

description of the data access path. Describes the data access path. The DMR must is able to describe not only the data model but also the data access path, i.e., mapping to the caches and SE RVers where data can be retrieved. DMR must be able to describe metadata models and data access paths, such as mapping data caches and servers.

ability to span applications. Cross application capabilities. The DMR is isn't focused on a single application or usage context, such as customer data, but spans all types the custom and Packaged applications and master data. DMR is not only concerned with an application or up and down related applications (such as customer data), but also extends to all types of custom/packaged applications and master data.

Original: Information fabric:enterprise Data Virtualization Download