Browse Prior Art Database

Verification of Data Integrity in a One-to-Many Piping Environment

IP.com Disclosure Number: IPCOM000104708D
Original Publication Date: 1993-May-01
Included in the Prior Art Database: 2005-Mar-19
Document File: 2 page(s) / 72K

Publishing Venue

IBM

Related People

Loveland, SW: AUTHOR

Abstract

Disclosed is a technique for continuously verifying that data records have not been lost or duplicated as they flow sequentially from a writer to several readers, where the readers perform destructive reads on the data and do not serialize between themselves on read operations. This technique can be used in testing computer applications with multiple readers.

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

Verification of Data Integrity in a One-to-Many Piping Environment

      Disclosed is a technique for continuously verifying that data
records have not been lost or duplicated as they flow sequentially
from a writer to several readers, where the readers perform
destructive reads on the data and do not serialize between themselves
on read operations.  This technique can be used in testing computer
applications with multiple readers.

      "Data streams" as discussed below refer to the following:  One
writer sends data records into the stream in sequential order.
Several readers read these records in a destructive and unpredictable
manner.  Destructive means that once a record has been read from the
stream by one reader, it disappears from the stream and is not
available to be read by other readers.  Unpredictable means that it
cannot be predicted in advance which records a given reader will read
from the stream.  However, regardless of which reader reads an
individual record, it is guaranteed that all records will be read
from the data stream in the same order they entered it.

      For each data stream through which data is flowing, a checklist
is maintained by the program in a location where it is accessible to
both the writer and readers.  The checklist contains two key pieces
of information for each record of data: 1) whether or not the record
has been read by anyone yet, 2) how many readers are PAST this record
in the data stream.  Also, part of the input to each reader is the
total number of readers participating in the test.  The writer sends
records to the data stream in sequential order with a sequence number
in each record to uniquely identify it (first record has a "1",
second record has a "2", etc).  When a reader reads a record, it does
the following:

1.  Mark it as read in the checklist.  If it is already marked as
    read then it has been duplicated, so the program terminates and
    provides appropriate diagnostic information.

2.  Inc...