How does a transaction processing system store and retrieve high volumes of transaction data reliably?
Describe storing and retrieving in transaction processing systems, including the data storage hierarchy, updating master files from transaction files, backup and recovery, and processing logs
A focused answer to the HSC Information Processes and Technology option dot point on storage and retrieval in transaction processing systems. Master and transaction files, updating, backup and recovery, and processing logs, with the traps markers look for.
Reviewed by: AI editorial process; not yet individually human-reviewed
Have a quick question? Jump to the Q&A page
What this dot point is asking
NESA wants you to describe how a transaction processing system (TPS) stores and retrieves its data given the high volumes and the need for reliability. You should explain master and transaction files and how one updates the other, and describe the backup, recovery and logging that protect a system the organisation cannot afford to lose. This builds on the characteristics of TPS dot point.
The answer
Master files and transaction files
A TPS distinguishes two kinds of data file. The master file holds the current, relatively permanent data, such as customer balances, product prices or stock on hand. The transaction file holds the recent activity, the individual sales, payments and adjustments, since the last update. Processing combines them: each transaction is applied to the master file to bring it up to date, after which the transaction file may be archived.
Updating the master file
In batch processing, transactions accumulate in the transaction file and are applied to the master file together in a scheduled run, often overnight. In real time processing, each transaction updates the master file the moment it occurs, so the master file is always current. The choice affects how fresh the stored data is and how heavy the processing load is at any moment; the characteristics dot point compares the two in full.
The storage hierarchy
A TPS uses different storage for different jobs. Fast online storage (solid state and hard disk, live database servers) holds the active master file so transactions can be applied quickly. Slower, cheaper media (high capacity disk, the cloud, tape) hold backups and archived transaction files where instant access is not needed. Matching the storage to the access need keeps the system both fast and affordable.
Backup and recovery
Because an organisation depends on its TPS, losing the master file could halt the business. The system therefore takes regular backups, copying the master file to separate, often offsite, storage. Recovery procedures define how to restore service after a failure: typically, restore the most recent backup and then re-apply every transaction recorded since that backup, returning the master file to its state at the moment of failure.
Processing logs
A processing log (or transaction log) records every transaction the system processes, in order, as it happens. The log serves recovery by providing the list of transactions to replay after restoring a backup. It also supports auditing, letting the organisation trace exactly what happened and when, and it underpins data integrity by giving an authoritative record independent of the master file.
Retrieving data
Retrieval in a TPS must be fast and reliable, because operators and customers query current data constantly (a balance, a stock level, an order status). Indexes on key fields speed these lookups, and the live master file in online storage ensures the answer reflects all processed transactions. The reliability of retrieval is part of why data integrity and concurrency control matter so much in this option.
Exam-style practice questions
Practice questions written in the style of NESA exam questions on this dot point, with worked answer explainers. The year tag is the paper they imitate, not the source.
2019 HSC4 marksA rental agency performs a nightly partial backup and a weekly full backup of its data. Explain the benefits to the rental agency of using a combination of partial backup and full backup.Show worked answer β
For 4 marks explain what each backup type does and why combining them benefits the agency.
Full backup (weekly). Copies all of the agency's data. It gives a complete restore point but takes a long time and a lot of storage, so doing it only weekly is practical.
Partial (incremental) backup (nightly). Copies only the data that changed since the last backup. It is fast and uses little storage, so it can run every night with minimal disruption.
Benefit of combining. The agency gets the best of both - frequent nightly partial backups limit how much recent work is lost if data is corrupted (good recovery point), while the weekly full backup provides a complete baseline to restore from. Together they balance speed, storage cost and data safety.
Markers reward defining both types and explaining how the combination improves recovery while keeping backups efficient.
2020 HSC4 marksA medical centre maintains patient records electronically and updates each record after every visit. Discuss the use of the grandfather, father, son back-up procedure in this medical centre.Show worked answer β
"Discuss" at 4 marks wants how the procedure works plus its advantages and limitations for the medical centre.
How it works: three generations of backup are kept. The most recent is the "son," the previous is the "father," and the oldest is the "grandfather." Each new backup becomes the son, the old son becomes the father, and so on, with the oldest cycled out.
Advantages: if the current data is corrupted, the centre can roll back to an earlier intact generation, so patient records can be recovered. Keeping three versions protects against a fault that is only discovered after a backup has already overwritten one copy.
Limitations: it uses more storage media, and because backups are periodic, any changes made since the last backup (for example today's consultations) could still be lost. For sensitive medical data, security of the stored copies is also a concern.
Markers reward explaining the rotation of three generations and weighing recovery benefit against storage and data-loss limits.
2021 HSC3 marksA shoe store needs to backup its sales data daily. Describe TWO forms of storage media suitable for backing up this system.Show worked answer β
For 3 marks describe two suitable backup media and why each fits a daily store backup.
External hard disk drive (or SSD). High capacity and fast read/write, so the day's sales data is backed up quickly. It can be disconnected and stored offsite for safety. Suitable for regular, sizeable daily backups.
Cloud / online storage. Data is copied over the internet to a remote data centre. It is automatic, offsite by design (protected against fire or theft at the store), and easily scalable, though it depends on internet access.
Magnetic tape is another valid answer (cheap, high capacity, good for archival backups). Markers reward two named media with a justification of capacity, speed or offsite safety for backing up daily sales data.
2022 HSC6 marksExplain the use of mirroring and rollback procedures in maintaining both data security and data integrity.Show worked answer β
For 6 marks explain each procedure and link it to both data security and data integrity.
Mirroring: data is written simultaneously to two (or more) storage devices that hold identical copies.
Data security: if one disk fails, the mirror still holds the data, so it is not lost - this protects availability.
Data integrity: both copies stay consistent and current, so the system always has an accurate, complete set of data to work from.
Rollback: if a transaction fails part-way or an error occurs, the system uses a log to undo the incomplete changes and return the database to its last consistent state.
Data integrity: it prevents partial or corrupt transactions from leaving the data in an invalid state, keeping records accurate and complete.
Data security: combined with the transaction log, it allows recovery to a known good point after a failure or attack.
Conclude that mirroring guards against media failure while rollback guards against transaction failure; together they protect both the safety and accuracy of the data. Markers reward defining each procedure and explicitly tying both to security and integrity.