Preparing your result...
Loading...
Press Esc to dismiss this message

Performance Optimization for CIM association traversal for WMI Providers (21-May-2008)

Thumbnail
IP.com Prior Art Database Disclosure (Source: IPCOM)
Disclosure Number IPCOM000170589D dated 21-May-2008
Originally published in Prior Art Database
Disclosed by: IBM
Country: Undisclosed
Disclosure File: 2 pages / 87.9 KB / English (United States)

Disclosed is a technique to improve the performance of an association call by supporting association methods directly. Association related methods are made extrinsic and call to these made using invoke method.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 61% of the total text.

Page 1 of 2

Performance Optimization for CIM association traversal for WMI Providers

Disclosed is a technique to improve the performance of an association call by supporting association methods directly. Association related methods are made extrinsic and call to these made using invoke method.

Current design of the associator call:


1.

2.

CLIENT

Associators()

Request/response

WMI to CMPI Converter

GetInstance()

Request/response

 EnumInstance() Request/response

GetInstance()

Request/response

WMI - CIMOM

GetInstance()

Request/response

 EnumInstance() Request/response

GetInstance()

Request/response

WMI to CMPI Converter

GetInstance()

Request/response

 EnumInstance() Request/response

GetInstance()

Request/response

CMPI PROVIDER

Client makes a call associators() by passing Source COP, Association Class Name and

Target Class Name.

WMI to CMPI Converter identifies it as associators() CIM call and since WMI does not

support associators() breaks the call into three different supported CIM calls and executes in sequence.
1. getInstance() on the Source COP
2. enumerateInstance() on Association Class Name
3. getInstance() on the matching Target COP. If there N matching target COP then there will be N getInstance calls.

Basically associators() method implemented by the provider never gets called. Instead algorithm for the same is placed in the WMI to CMPI Converter. With this current approach performance is in Order(n) compared to the new, improved appro...

(Source: IPCOM)
First page image
(Source: IPCOM)