ISYS1055 Chap.1 Databases vs File-Based Storage: What a DBMS Adds
Databases vs File-Based Storage: What a DBMS Adds
Week 1 of RMIT University ISYS1055 Database Concepts builds the whole course out of one argument: take a business that keeps its records in flat files or spreadsheets, list everything that goes wrong, and derive a database feature from each failure. The stated outcomes are to explore the features of relational database management systems, describe the limitations of traditional data storage, survey the DBMSs currently available, and distinguish relational engines (Oracle, SQL Server, MySQL, SQLite) from non-relational ones (MongoDB). This is the chapter behind the "why a database" style quiz items, and behind Part A of the Database Design Project, where you have to look at a real public dataset and say what it actually contains before you model anything.
What this chapter covers
- 01Data vs information; a database as a shared, self-describing, integrated collection of persistent, logically related data
- 02The DBMS as the layer between users and stored data - definition, construction, manipulation, sharing, protection; the working test "if it doesn't CRUD, it's not a database"
- 03CRUD and its SQL statements: INSERT, SELECT, UPDATE, DELETE; schema vs instance
- 04The DBMS sub-languages: DDL (CREATE / ALTER / DROP), DML (SELECT / INSERT / UPDATE / DELETE), plus DCL and transaction control
- 05The seven failures of file-based storage - queries, integrity, restructuring, multi-user access, crashes, security, scale - and the database concept each one motivates
- 06Transactions and ACID: atomic, consistent, isolated, durable; commit and rollback as the answer to the crash case
- 07Database security in layers: network, access management (authentication, authorisation, access control), threat protection, information protection
- 08The DBMS landscape - relational engines vs the document model - and why this course runs on SQLite through SQLite Studio, with LucidChart for diagrams and MongoDB Atlas from Week 11
Diagnose four spreadsheet failures and name the database concept each one motivates
- +1(i) is the integrity and consistency failure: nothing in a spreadsheet stops a mistyped value creating a phantom second person, so one borrower is now two and every count derived from the file is wrong. The fix is integrity constraints - a key constraint making the borrower identifier unique, and a foreign key so a loan can only reference a borrower who already exists.
- +1(ii) is the multi-user access failure - the classic lost update, where two writers load, edit and save and the later save destroys the earlier one. The fix is concurrency control inside a DBMS, which serialises conflicting work so that concurrent transactions do not overwrite each other's results.
- +1(iii) is the crash failure: a multi-step change is left half applied, so the data is now in a state the business never intended. The fix is the transaction - a logical unit of work that is atomic (all or none), consistent, isolated and durable - together with commit, rollback and recovery. Nineteen committed changes out of forty is precisely what atomicity forbids.
- +1(iv) is the query failure: text search over a file can find a name, but there is no execution path for a question that relates one set of rows to another. The fix is a schema plus a declarative query language - SQL as the DML - which lets you state what you want rather than how to compute it. As the collection grows, the same DBMS also supplies indexes and query optimisation, which is the separate efficiency and scale argument.
Key terms
- Database
- A shared, self-describing, integrated collection of persistent, logically related data, organised so that it can be efficiently stored, retrieved and maintained. The description of the structure (the schema) is held with the data, which is what makes it self-describing.
- DBMS
- The software layer sitting between users or applications and the stored data. It provides definition, construction, manipulation, sharing, protection and maintenance of databases. The working test used in this course is that if it does not support all four CRUD operations, it is not a database.
- CRUD
- Create, Read, Update, Delete - the four fundamental data operations, realised in SQL as INSERT, SELECT, UPDATE and DELETE. Weeks 5 to 8 concentrate on Read.
- DDL and DML
- The Data Definition Language defines and alters the schema (CREATE TABLE, ALTER TABLE, DROP TABLE, CREATE VIEW, CREATE TRIGGER); the Data Manipulation Language queries and changes the data (SELECT, INSERT, UPDATE, DELETE). Privileges (GRANT, REVOKE) and transaction boundaries (COMMIT, ROLLBACK) sit in their own sub-languages.
- Transaction and ACID
- A logical unit of work that must be Atomic (completes entirely or has no effect), Consistent (leaves the database satisfying its constraints), Isolated (unaffected by other concurrent work) and Durable (survives failure once committed).
- Schema vs instance
- The schema is the description of the structure - relations, attributes, domains, constraints - and changes rarely. The instance (or state) is the data held at a point in time and changes constantly; a DBMS normally keeps only the current instance.
Databases vs File-Based Storage: What a DBMS Adds FAQ
Why does this course use SQLite rather than a big server engine?
SQLite is a very light-weight, single-user, embedded database, it is open source, and it is the most widely deployed database in the world - it ships inside mobile operating systems and countless applications. That makes it installable on a laptop for the practical with no server administration, which matters because you have to bring a laptop to your practical class. The GUI used with it is SQLite Studio. The trade-off is that SQLite relaxes a few standard behaviours, and the course flags those explicitly when they arrive (foreign keys off by default in Week 2, and the GROUP BY deviation in Week 7).
What is the difference between a database and a DBMS?
The database is the data plus its self-describing structure; the DBMS is the software that defines, builds, manipulates, shares, protects and maintains it. When you connect SQLite Studio to a .db file you are running a DBMS over a database. Most of what this course teaches you to value - constraints, concurrency, transactions, views, query optimisation - is supplied by the DBMS, not by the file.
Is a transaction needed every time I run a query?
This is the discussion question the course poses in the first lectorial. Every statement executes inside a transaction, so a single read runs in an implicit one; you rarely think about it. Explicit transaction boundaries matter when one logical unit of work spans several statements - the half-finished batch of updates is exactly the case where atomicity has to hold across statements rather than within one.
Can AI help me with the Week 1 material in ISYS1055?
Yes, as a study aid. Sia is an AI tutor built to mirror how ISYS1055 is taught and assessed at RMIT University: it can drill you on the file-storage failure list, explain ACID with a worked scenario, or check the requirements list you drafted for a public dataset before you turn it into a model. It does not complete graded assessment for you, and RMIT University academic-integrity rules apply - confirm what is permitted on Canvas.
Assessment move
Do the two practical set-up jobs in Week 1 rather than in Week 5: install SQLite Studio, and connect it to a supplied .db file through Database > Add a Database so you know the Structure, Data and DDL tabs before you need them under time pressure. Then learn the seven-failure list as symptom-to-concept pairs, because that is the shape the first quiz uses - it runs in your practical in Week 3 and covers Weeks 1 and 2. Practise the requirements-elicitation move on any public website you use: name the classes of things its database must hold, the details held for each, and one question the site clearly cannot answer. That is the same skill Part A of the Database Design Project asks for, and Part A carries zero marks but gates everything after it.
Working through Databases vs File-Based Storage: What a DBMS Adds in ISYS1055? Sia is AskSia’s AI Information Technology tutor — ask any ISYS1055 Databases vs File-Based Storage: What a DBMS Adds question and get a clear, step-by-step explanation grounded in how ISYS1055 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
