Glossary · 72 terms

Cloud & DevOps glossary

Every new word in the Cloud & DevOps course, in plain English, explained on a card the first time it comes up.

A

agent
a program that acts on its own in a loop; a GitOps agent keeps the cluster matching what Git saysUnit: Run it reliably
API
a defined way for one program to ask another program for data or actionsUnit: Cloud and DevOps from zero
API call
one request sent to an API, and the response it returnsUnit: Release and observe
attacker
a person deliberately trying to break into or misuse a systemUnit: Run it reliably

B

backend
the system that receives and stores data; for telemetry, the tool where traces and metrics end upUnit: Release and observe
backup
a copy of data kept so it can be restored after loss or damageUnit: Run it reliably
batching
grouping many items so they are handled together instead of one by oneUnit: Cloud and DevOps from zero
billing
working out and sending what each customer owesUnit: Cloud and DevOps from zero
blast radius
how much of the system or how many users a failure can affectUnit: Run it reliably
bottleneck
the one part of a system that limits how fast the whole thing can goUnit: Release and observe
branch
a separate line of work in Git, merged back when doneUnit: Cloud and DevOps from zero
byte
the basic unit of computer data; a gigabyte is about a billion bytesUnit: Run it reliably

C

chaos engineering
testing a system by causing controlled failures to learn how it behavesUnit: Run it reliably
chargeback
billing each team's cloud spend to its own budgetUnit: Run it reliably
CPU
the main processor chip that runs a computer's programsUnit: Release and observe
credentials
the secret details, such as a key or password, that prove who is askingUnit: Cloud and DevOps from zero

D

dashboard
a screen of charts and numbers that shows how things are going at a glanceUnit: Release and observe
database
an organised store of data that programs can save to and look things up inUnit: Containers, pipelines and infrastructure as code
dependency
something your work relies on, such as a library your code uses or a task that must finish firstUnit: Containers, pipelines and infrastructure as code
deploy
to put a new version of software live where users can reach itUnit: Cloud and DevOps from zero
Dockerfile instructions
FROM picks the base image, COPY adds your files, RUN executes a command while building, CMD sets what startsUnit: Containers, pipelines and infrastructure as code

E

encryption
scrambling data so only someone with the right key can read itUnit: Containers, pipelines and infrastructure as code

G

golden signals
latency, errors, traffic and saturation: four signals to watch for every serviceUnit: Release and observe

H

headroom
spare capacity above the expected peakUnit: Release and observe

I

incident
an event that disrupts a service or puts data at risk and needs a responseUnit: Cloud and DevOps from zero
Ingress
rules that route outside web traffic to Services by host or pathUnit: Containers, pipelines and infrastructure as code
inject
to insert something into a running system from outsideUnit: Containers, pipelines and infrastructure as code

L

latency
how long one request takes to get an answerUnit: Release and observe
layer
one saved step of an image build that can be reused from cacheUnit: Containers, pipelines and infrastructure as code
lead time
how long a change takes from code committed to running in productionUnit: Run it reliably
least privilege
giving only the access a job needsUnit: Cloud and DevOps from zero
load balancer
a service that spreads incoming requests across several serversUnit: Cloud and DevOps from zero
load test
sending controlled, realistic traffic at a system to measure how it behaves under loadUnit: Release and observe
log
a timestamped record a program writes about what it didUnit: Containers, pipelines and infrastructure as code

M

margin
a safety buffer left between where you are and a limitUnit: Release and observe
merge
to combine two things into one, like joining a code branch back into mainUnit: Cloud and DevOps from zero
multi-stage build
a Dockerfile with a build stage and a slim final stage that copies only the outputUnit: Containers, pipelines and infrastructure as code

O

on-call
the engineer currently responsible for responding when alerts page themUnit: Cloud and DevOps from zero
outage
a period when a service is down or not workingUnit: Run it reliably

P

patching
installing updates that fix flawsUnit: Cloud and DevOps from zero
payroll
paying employees' wagesUnit: Cloud and DevOps from zero
pipe
the | symbol, which sends one command's output into anotherUnit: Containers, pipelines and infrastructure as code
pod
the smallest unit Kubernetes runs: one or more containers togetherUnit: Containers, pipelines and infrastructure as code
policy
a written list of allowed actions on named resourcesUnit: Cloud and DevOps from zero
production
the live system real users use, as opposed to test copiesUnit: Cloud and DevOps from zero
pull request
a proposed code change that teammates review before it is mergedUnit: Containers, pipelines and infrastructure as code

Q

queue
a line of work waiting its turnUnit: Release and observe

R

README
the introduction file at the top of a code project that explains what it is and how to use itUnit: Containers, pipelines and infrastructure as code
refactor
restructure code without changing what it doesUnit: Run it reliably
replication
keeping copies (replicas) of data or services on several machinesUnit: Containers, pipelines and infrastructure as code
repository (repo)
the place where a project's code and its history are storedUnit: Containers, pipelines and infrastructure as code
responsibility model
the agreed split of which party looks after which part of a serviceUnit: Cloud and DevOps from zero
revert
undo a change by going back to the earlier versionUnit: Run it reliably
role
an identity a service takes on for a while, with temporary credentialsUnit: Cloud and DevOps from zero
rollback
going back to the previous working version after a bad changeUnit: Cloud and DevOps from zero
rollout
releasing a change to users, often step by stepUnit: Release and observe
route
the path data or traffic travels alongUnit: Release and observe
runbook
a short, ordered list of checks and actions for a known alert or taskUnit: Run it reliably
runtime
the moment a program is actually running, or the software that runs itUnit: Containers, pipelines and infrastructure as code

S

script
a small program that automates a taskUnit: Checkpoint: Cloud basics
Service
a stable address that spreads traffic across a set of podsUnit: Containers, pipelines and infrastructure as code
shell
a text interface for running programs and commandsUnit: Containers, pipelines and infrastructure as code
showback
reporting each team's cloud spend to them without billing itUnit: Run it reliably
sprint
a fixed one-to-two-week work cycle in agile teamsUnit: Run it reliably
SRE (site reliability engineering)
running production systems with engineering methods: automation, SLOs and error budgetsUnit: Run it reliably
stakeholder
anyone affected by or with a say in a piece of workUnit: Run it reliably
subnet
a slice of a network's addresses, often marked public or privateUnit: Cloud and DevOps from zero
sync
to make two things match; a sync agent keeps the cluster matching GitUnit: Run it reliably

U

UI (user interface)
the screens, buttons and text a person uses to work an appUnit: Checkpoint: Run it reliably
uptime
the share of time a service is working; downtime is when it isn'tUnit: Run it reliably
USE method
checking utilisation, saturation and errors for every resourceUnit: Release and observe

V

virtual network
a private network you define inside a cloud providerUnit: Cloud and DevOps from zero

See what the Cloud & DevOps course covers

Your first Cloud & DevOps lesson is five minutes away.

Not in the stores yet. Leave your email and we will tell you when it launches.

CEO Hero: launching soon

CEO Hero is not in the app stores yet. Leave your email for one note when it launches on Android.

Used only to tell you once when it launches.