| 摘要 |
PURPOSE:To utilize effectively capacity of a file, by deciding automatically size of an optimum idle area from record length of a record which is written actually. CONSTITUTION:A record is stored fixedly, for instance, by 80%, and the remaining 20% is used as an idle area, and as a result, mean record length l1 of its block is derived. Subsequently, l1 is regarded as mean record length of this block, and an idle record area E2 of this block is forecast. Subsequently, the record is stored until exceeding (L-E2) (L : maximu data quantity which can be stored in the block), and as to the record of that time and before, mean record length l2 of this block is derived, and an idle area E'2 of this block is calculated. In case of E2-E'2<=0, the processing is shifted to the following block, and in case of E2-E'2>0, the record is stored until exceeding (L-E'2). |
