4 Architecting IoT Cloud
195
• Record: Records are also known as rows of data which are horizontal entities of
a table. For each entry in the table, there is one row in the table.
• Field: Tables are made up of smaller entities known as fields; these are columns
in a table that hold unique information regarding each table record.
• Key: A key can uniquely identify a row in the table. Keys also enable you to
establish a relationship between tables.
• Database normalization: Database normalization is the means for efficiently
organizing data within a database in order to (1) remove unnecessary data such
as duplicate data across tables and (2) guarantee that data dependencies in
database are properly enforced. These goals are important because they decrease
the amount of space a database requires while ensuring logical data storage.
Normalization guidelines are helpful in developing a solid database structure.
In general, there are three major normal forms in a database:
– 1st normal form – There are no repeating groups of columns in a table.
– 2nd normal form – The table is in first normal form and all the columns are
fully functional dependent on the table’s primary key.
– 3rd normal form – A table in third normal form is a table in the second normal
form that has no transitive dependencies. A transitive dependency can occur
when: P - > Q and Q - > R is true, then P- > R is a transitive dependency. Note
that P, Q, and R are attributes (columns) of a table.
• ACID compliance: ACID compliance in transactions is one of the most important
features of RDBMS. A transaction is a logical unit that typically consists of
several low-level tasks which should be executed for data retrieval or updates
in a database. A transaction must maintain atomicity, consistency, isolation, and
durability (known as ACID) to guarantee the accuracy and completeness, as well
as the integrity of data [16].
– Atomicity – This property makes sure that all of a transaction’s operations
are executed or none are executed; partial completion of a transaction is not
tolerated.
– Consistency – The database must remain in a stable state before and after a
transaction is completed. Transactions should not have a negative effect on
data stored in the database.
– Durability – The database must maintain all recent updates in the event that a
system fails or restarts. If a transaction updates a portion of data and commits
to it, then the database has to maintain the updated data. If a system fails
before a committed transaction can write the data to the disk, then the data
update must take place when the system becomes functional again.
– Isolation – If multiple transactions are being executed at the same time and in
parallel, then all transactions will be completed as if it was the only system
transaction. No single transaction affects any other transaction.
195
• Record: Records are also known as rows of data which are horizontal entities of
a table. For each entry in the table, there is one row in the table.
• Field: Tables are made up of smaller entities known as fields; these are columns
in a table that hold unique information regarding each table record.
• Key: A key can uniquely identify a row in the table. Keys also enable you to
establish a relationship between tables.
• Database normalization: Database normalization is the means for efficiently
organizing data within a database in order to (1) remove unnecessary data such
as duplicate data across tables and (2) guarantee that data dependencies in
database are properly enforced. These goals are important because they decrease
the amount of space a database requires while ensuring logical data storage.
Normalization guidelines are helpful in developing a solid database structure.
In general, there are three major normal forms in a database:
– 1st normal form – There are no repeating groups of columns in a table.
– 2nd normal form – The table is in first normal form and all the columns are
fully functional dependent on the table’s primary key.
– 3rd normal form – A table in third normal form is a table in the second normal
form that has no transitive dependencies. A transitive dependency can occur
when: P - > Q and Q - > R is true, then P- > R is a transitive dependency. Note
that P, Q, and R are attributes (columns) of a table.
• ACID compliance: ACID compliance in transactions is one of the most important
features of RDBMS. A transaction is a logical unit that typically consists of
several low-level tasks which should be executed for data retrieval or updates
in a database. A transaction must maintain atomicity, consistency, isolation, and
durability (known as ACID) to guarantee the accuracy and completeness, as well
as the integrity of data [16].
– Atomicity – This property makes sure that all of a transaction’s operations
are executed or none are executed; partial completion of a transaction is not
tolerated.
– Consistency – The database must remain in a stable state before and after a
transaction is completed. Transactions should not have a negative effect on
data stored in the database.
– Durability – The database must maintain all recent updates in the event that a
system fails or restarts. If a transaction updates a portion of data and commits
to it, then the database has to maintain the updated data. If a system fails
before a committed transaction can write the data to the disk, then the data
update must take place when the system becomes functional again.
– Isolation – If multiple transactions are being executed at the same time and in
parallel, then all transactions will be completed as if it was the only system
transaction. No single transaction affects any other transaction.