The patent's assignee for patent number 8768959 is
News editors obtained the following quote from the background information supplied by the inventors: "Relational Database Management Systems are sometimes referred to as RDBMS.
"Distributed RDBMS that are implemented according to a 'Shared Nothing' system architecture are sometimes referred to as Parallel Database Servers (in contrast to Symmetric Multiprocessing (SMP) Database Servers) and when the number of logical computers in the Parallel Database Server is large, this is sometimes referred to as 'Massively Parallel Processing' or MPP architecture.
"Conventionally, the number of logical computer servers that participate in parallel database architectures remains constant during the life of the system. Elastic Parallel Database Servers are an extension of the static MPP architecture where the number and composition of the computer servers may change during query processing, often in response to the demands being placed on the system. We refer to Elastic Parallel Database Management Systems as EPRDBMS."
As a supplement to the background information on this patent, VerticalNews correspondents also obtained the inventors' summary information for this patent: "Relational Database Management Systems (RDBMS) maintain metadata information about the schema, and in the case of EPRDBMS the metadata includes information about the placement of data on the system.
"In cases where an RDBMS is used to store data for a Software as a Service (SaaS) application, it is very common that each client of the service has its data and customizations stored in a database on the RDBMS. In this kind of application architecture, there is a one-to-one mapping between clients and databases, and therefore a SaaS application with many clients would necessarily have many databases.
"In an SMP RDBMS, each database resides on a single server and in an MPP RDBMS each database may reside on a plurality of database servers. For the purposes of this preceding statement we refer only to a single copy of the database, but multiple copies of the data may be maintained for the purpose of high-availability and redundancy.
"The Elastic Parallel RDBMS (EPRDBMS) includes a Dynamic Query Planner that converts queries submitted to the application into query execution plans that consist of a sequence of operations that must be performed by individual nodes in the system in a specified order in order to produce the results expected by the application.
"Data in the EPRDBMS is stored in tables that are associated with storage group(s), and the data in the tables is distributed across a plurality of storage nodes that are members of the storage group, using one of many distribution methods, some of which are data dependent and some of which are data independent. Tables are grouped into logical collections called databases.
"Elasticity in the EPRDBMS is achieved through abstractions like storage nodes grouped into storage nodegroups, and compute nodes grouped into compute nodegroups, and mechanisms that allow for these to be provisioned dynamically based on the workload being placed on the system at any given instant of time.
"The EPRDBMS ensures that applications are guaranteed well understood database ACID properties; Atomicity, Consistency, Isolation, and Durability, and it also provides applications with the ability to perform groups of operations that are either all completed or all aborted but never a situation where some but not all of the operations are completed.
"Unless explicitly described to the contrary, this document only refers to the first (primary) copy of a database. Without loss of generality, concepts described here also apply to all copies of the database. This is for simplicity of illustration and so as to not obscure key aspects of the invention.
"Commands and operations are submitted to a database in some query language that is mutually understood by the client and the EPRDBMS, such as the Structured Query Language (SQL) on a `connection` established by the client with the EPRDBMS for this purpose. Persistent state about the connection is maintained by the database along with each connection."
For additional information on this patent, see: Annapragada, Mrithyunjaya; Rousseau, Benjamin. Dynamic Rules Based Query Rewriting Engine. U.S. Patent Number 8768959, filed
Keywords for this news article include:
Our reports deliver fact-based news of research and discoveries from around the world. Copyright 2014, NewsRx LLC
Most Popular Stories
- Businesses, Investors Pressing for Green Policy
- Who's Next? More Nude Celeb Pics Hacked, Leaked
- Tips for Hiding, Securing Data on Smartphones
- Hispanic Enterprises Drive U.S. Economy
- ISIS Calls for Jihad Against 'Filthy French'
- Fed in No Rush to Raise Interest Rates
- Would You Trade Privacy for Job Security?
- Cristela Gets a Big Thumbs Up
- Lower Used-Car Prices Roil the Auto Industry
- Iran Says Syria Strikes Illegal