Glossary · 55 terms
Data & Databases glossary
Every new word in the Data & Databases course, in plain English, explained on a card the first time it comes up.
A
- admin rights
- full control over a system: the most powerful and risky kind of accessUnit: Pipelines and analytics
- aggregation
- combining many rows into one number, such as a count, sum or average per groupUnit: Model and query
- alert
- an automatic message that tells people something needs attention nowUnit: Pipelines and analytics
- atomic
- all-or-nothing: the steps happen together or not at all, and nothing can sneak in betweenUnit: Scale and recover
- audit log
- a permanent record of who did what and when, kept so it can be checked laterUnit: Pipelines and analytics
- availability zone
- a separate data centre within a cloud region, so one failing doesn't take down the othersUnit: Scale and recover
B
- batching
- grouping many items so they are handled together instead of one by oneUnit: Model and query
- byte
- the basic unit of computer data; a gigabyte is about a billion bytesUnit: Pipelines and analytics
C
- cache
- a fast store of recent results, so repeated requests don't redo the workUnit: Scale and recover
- cardinality
- how many of one thing can be linked to how many of anotherUnit: Model and query
- CHECK
- a constraint that rejects values failing a conditionUnit: Model and query
- cluster
- a group of machines or items that work or sit togetherUnit: Scale and recover
- consensus
- a protocol that makes a majority of copies agree before a write countsUnit: Scale and recover
- constraint
- a rule on a column or table that the database enforces on every writeUnit: Model and query
- CPU
- the main processor chip that runs a computer's programsUnit: Scale and recover
- CTE
- common table expression: a named sub-query defined with WITHUnit: Pipelines and analytics
D
- DAG
- directed acyclic graph: steps linked one way, with no loopsUnit: Pipelines and analytics
- dashboard
- a screen of charts and numbers that shows how things are going at a glanceUnit: Checkpoint: Model and query
- DBMS
- database management system: software that stores data and enforces its rulesUnit: Data from zero
- dependency
- something a step needs to finish before it can runUnit: Pipelines and analytics
- distributed SQL
- a SQL database that spreads data over many nodes while keeping transactionsUnit: Scale and recover
- document store
- a database that keeps whole JSON-like recordsUnit: Data from zero
- drift
- slowly moving out of line, so copies or results stop matchingUnit: Model and query
- duplicate rows / dedupe
- the same record stored more than once; deduping removes the extrasUnit: Model and query
E
- ER diagram
- a drawing of entities, their attributes and the relationships between themUnit: Model and query
- eviction
- removing an entry from the cache to make roomUnit: Scale and recover
I
- index
- a sorted lookup structure that lets a database or search engine find items without reading everythingUnit: Model and query
K
- key-value store
- a database that fetches one value by its keyUnit: Data from zero
L
- lineage
- a record of where a table's data came from and how it was changedUnit: Pipelines and analytics
- lock
- a database hold that stops others changing rows or a table until the current change finishesUnit: Model and query
- log
- a timestamped record a program writes about what it didUnit: Pipelines and analytics
- lookup
- finding one item directly by its keyUnit: Data from zero
M
- merge
- to combine two things into one, like joining a code branch back into mainUnit: Pipelines and analytics
- MVCC
- multi-version concurrency control: keeping several versions of a row so readers see a consistent snapshotUnit: Scale and recover
N
- nested
- placed inside another of the same kind, like a list inside a listUnit: Data from zero
O
- orchestrator
- a tool that runs pipeline steps in order, on a schedule, with retriesUnit: Pipelines and analytics
- outage
- a period when a service is down or not workingUnit: Model and query
P
- production
- the live system real users use, as opposed to test copiesUnit: Pipelines and analytics
Q
- quarantine
- set aside or isolate something so it can't affect the rest until it is checkedUnit: Scale and recover
- query
- a question asked of a database or search systemUnit: Data from zero
R
- replay
- re-running recorded events or history to rebuild the same stateUnit: Scale and recover
- replicate
- to keep copies of data on several machinesUnit: Scale and recover
- retry
- running a failed step again, usually after a short waitUnit: Pipelines and analytics
S
- SaaS
- software as a service: software you rent and use over the internetUnit: Pipelines and analytics
- scale out
- handle more load by adding more machines rather than a bigger oneUnit: Scale and recover
- scheduled job
- a task that runs automatically at set times, such as a nightly job at 2 amUnit: Pipelines and analytics
- script
- a small program that automates a taskUnit: Model and query
- server
- a computer that runs programs and answers requests from other machinesUnit: Data from zero
- snapshot
- a frozen copy of data at one momentUnit: Scale and recover
T
- timestamp
- the recorded date and time something happenedUnit: Scale and recover
- turnaround
- the ground time between an aircraft arriving and departing againUnit: Pipelines and analytics
U
- uptime
- the share of time a service is working; downtime is when it isn'tUnit: Scale and recover
W
- window function
- a calculation over related rows that keeps each row, written with OVERUnit: Pipelines and analytics
- workload
- the work a system has to run, such as customers' jobs or queriesUnit: Pipelines and analytics
- write-through
- a cache write that also writes the database at the same timeUnit: Scale and recover
Your first Data & Databases lesson is five minutes away.
Not in the stores yet. Leave your email and we will tell you when it launches.