Browse Prior Art Database

System for Memory Management in Multi-Engine Distributed Database Products

IP.com Disclosure Number: IPCOM000241993D
Publication Date: 2015-Jun-12
Document File: 4 page(s) / 132K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method to use a monitoring infrastructure and a self-aware self-correcting mechanism to minimize memory wastage in multi-engine products. The solution increases capacity and scalability by reducing memory waste.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 4

System for Memory Management in Multi-Engine Distributed Database Products


Information technology (IT) providers are faced with the problem of the addition of parallel or secondary engines to existing tuned engines that are characterized by disparate paradigms in terms of how the engines treat finite and scarce system resources , such as memory.

Unintentional memory wastage is one of the known pitfalls of otherwise efficient data systems . In such data systems, there are known cases of two independently developed data engines needing to work harmoniously in order to process in-memory data from, for example, row and column organized tables simultaneously.

A method is needed to ensure an improved inter-working and allow users of in-memory databases to increase the efficiency of available physical memory to run through more work in parallel.

The novel contribution is a method to use a monitoring infrastructure and a self -aware self-correcting mechanism to minimize memory wastage in multi-engine products. The solution increases capacity and scalability by reducing memory waste. Improved, seamless inter-operability between two parallel engines reduces Total Cost of Ownership. The method provides the flexibility to switch between resource-constrained systems vs. performance-intensive systems. The novel solution allows the co-habitation of disparate architectures and the associated commensurate demands on system resources , when multiple engines are engaged in addressing parallel homogeneous requests or divisible heterogeneous requests from the viewpoint of the algorithms from engines in question.

This novel approach enables the coexistence of disparate architectures, as opposed to the classic "rewrite one of them" approach. The rewrite typically solves the general problem, but at the expense of potentially breaking a component that was already working.

Additionally, the balancing of system resources allocation among two or more engines is pr...