Answer Posted / stu
Data is read and written to files in blocks. The reading
and writing of blocks is I/O. I/O is overhead (increases
runtime and cost) and you want to keep it to a minimum.
When reading a file, a whole block of records will be read,
but the records in the block will be fed to the program one
at a time until the block is exhausted and the next block is
read. The operating system does this behind the
scenes--it's transparent to the programmer.
When writing to a file, the records are not actually written
to the file until a block is full (or the program ends--at
which time the last block is written if it contains at least
1 record.) Again, this is transparent to the programmer.
Obviously, to keep I/O low, you want the cram as many
records into a block as possible.
Let's say your FB file's record length (LRECL) is 252 bytes
and your facility allows block sizes up to 27000 bytes.
Your blocksize would be ((integer (27000 / 250)) * 252) or
26964 (107 records).
| Is This Answer Correct ? | 0 Yes | 0 No |
Post New Answer View All Answers
I have a COBOL program that ACCEPTs some input data. How do you code the JCL statement for this?
What is the function of the dd mgmtclas keyword in sms datasets?
List the various advantages of using jcl language?
by using cond parameter maximum 8 cond can be coded in single cond means ?explain
Is condition checking possible in jcl?
what is “Cond= even” and “Cond=only”?
What is one line to pass PARM from JCL to COBOL?
Name the statement which can be used to send data to another mvs jes3 node?
How to override loadlib?
how you can access an uncataloged dataset in a JCL?
List in order the hierarchical levels of jcl?
what is use of disp parameter in dd statement?
Explain the function of //jcllib statement?
How to alter the parameters for the existing gdg?
Is it possible to define dd statements as you want?
