In this recipe, we will be discussing sequential scans.
Sequential scans are a mechanism, and PostgreSQL tries to read each tuple from the relation. The best example for the sequential scan is reading an entire table without any predicate. Sequential scans are always preferred over index scans, when a query is reading most of the data from the table, which will avoid the index lookup overhead.
Reading pages in sequential order takes less effort when compared with reading pages in random order. This is because, in sequential file reading, we do not need to set the file pointer to any specific location. However, during the index scan, PostgreSQL needs to read random pages from the file as per the index results. That is, during the index scan we move the file read pointer to multiple pages. This is the reason why the arbitrary cost parameter seq_page_cost value 1 is always less than the random_page_cost value 4.