Browse Prior Art Database

Accumulating Performance Monitor for Printer Object Data Fetches in an FPGA

IP.com Disclosure Number: IPCOM000014852D
Original Publication Date: 2002-Apr-08
Included in the Prior Art Database: 2003-Jun-20
Document File: 2 page(s) / 47K

Publishing Venue

IBM

Abstract

The total cumulative time from "Request for New Object" to Receiving initial data is timed for both "Linework" (LW) and "Linework Control" LWC data for multiple objects. The number of both types of objects processed is counted as well. The average start-up time can then be calculated for the period of time required to get initial data for each decompressor and data type. This calculation is done for each decompressor in a processor after the print is complete. The average for the set of decompressors as a whole is also computed. The accumulating timer counts whenever the object startup monitor state machine indicates that a new object fetch has started and I have not yet gotten data for both the LW and LWC decompressors. This allows evaluation of the overall bus performance and the memory pipeline performance for new object fetches. The range that is evaluated can be limited/constrained in various ways. For instance, these numbers can be checked for just the sheet to sheet transition reload period by only counting objects and time elapsed during the time that the Transport address is less than a specified number. The design can also be modified to monitor performance within a specified sheet count range.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 87% of the total text.

Page 1 of 2

Accumulating Performance Monitor for Printer Object Data Fetches in an FPGA

The total cumulative time from "Request for New Object" to Receiving initial data is timed for both "Linework" (LW) and "Linework Control" LWC data for multiple objects. The number of both types of objects processed is counted as well.

The average start-up time can then be calculated for the period of time required to get initial data for each decompressor and data type. This calculation is done for each decompressor in a processor after the print is complete. The average for the set of decompressors as a whole is also computed.

The accumulating timer counts whenever the object startup monitor state machine indicates that a new object fetch has started and I have not yet gotten data for both the LW and LWC decompressors.

This allows evaluation of the overall bus performance and the memory pipeline performance for new object fetches. The range that is evaluated can be limited/constrained in various ways. For instance, these numbers can be checked for just the sheet to sheet transition reload period by only counting objects and time elapsed during the time that the Transport address is less than a specified number. The design can also be modified to monitor performance within a specified sheet count range.

This invention is composed of these pieces:

1) The Start-up monitor State Machine that has 4 states

0 idle

1 New Object has been requested

2 LW data has been requested

3 LWC data has been...