Browse Prior Art Database

Record Level Change Tracking Included within a Data Set Disclosure Number: IPCOM000239805D
Publication Date: 2014-Dec-02
Document File: 2 page(s) / 34K

Publishing Venue

The Prior Art Database


Disclosed is a method to update the metadata information for existing extended format data sets, such that corrupted data sets are easily detected.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 2

Record Level Change Tracking Included within a Data Set

Broken data sets can occur for many different reasons. A few examples include but are not limited to software program

error, serialization errors, and hardware failures. Diagnosing the cause of a broken data set is often difficult and time consuming. When a block of data is written, the entire block is written to a Direct Access Storage Device (DASD) at one time. In some cases, an entire track of blocks is written at one time. Many times, only a small portion of the block is actually bad.

One type of diagnosis method is tracing these blocks as the blocks are written. This trace data is then written to an external data set and the input/output (I/O) is traced as the entire block is written. This requires the client to have a tracing facility such as Generalized Tracing Facility (GTF) active and set to this particular device. A better methodology is

needed in the area of diagnosing corrupted data sets.

Today on a 64-bit operating system (OS) for mainframe computers, extended format data set types already have a small portion of the data set imbedded with metadata information. This information may include items like the current block count. It can be updated to include information related to the last time this block changed. Thus, information required to diagnose erroneous data can be included within the data set, in the metadata region, or prefix which resides on the same track as the data.

The novel contribution is a method to update the metadata information for existing extended format data sets, which

today use a small portion of a fixed 32-byte prefix, or add it where possible to non-extended format data sets to create a new metadata area within the data. For most data set block sizes, some extra space still exists within the track capacity for metadata. For example, in a 3390 Mode for equal length physical records without keys (non-Indexed Sequential

Access Method/Basic Sequential Access Method (ISAM/BSAM) with a data length of 3769-4136, 12 of these data blocks can fit per track. For a data set with a 4096 size block, 40 more bytes of information can be added without actually using any additional tracks on the DASD. Other block sizes have different amounts of unutilized space. This space can be used to tr...