Learn.
Cloud concepts, explained.
Storage, Kubernetes, network and operations concepts in plain terms: what each one is, how it works and how enum runs it.
Storage
What is the 3-2-1 backup rule?
The 3-2-1 backup rule says to keep three copies of important data, on two different types of storage, with one copy kept off-site.
Read the articleWhat is synchronous replication?
Synchronous replication confirms a write only after every copy has stored it, so a failed server or disk loses no acknowledged data.
Read the articleWhat is an immutable backup?
An immutable backup is a backup copy that nobody can change or delete until a set date, so ransomware, mistakes and stolen credentials cannot destroy it.
Read the articleWhat is an S3 lifecycle rule?
An S3 lifecycle rule tells a bucket to act on objects automatically once they reach a set age or date, such as deleting old logs or versions.
Read the articleWhat is data durability?
Data durability is the probability that stored data stays intact and is not lost over a year, usually stated in nines and separate from availability.
Read the articleWhat is S3 versioning?
S3 versioning keeps every version of an object in a bucket, so overwrites and deletes can be undone by restoring an earlier version.
Read the articleWhat is S3 Object Lock?
S3 Object Lock stores object versions as write-once-read-many (WORM), so a locked version cannot be overwritten or deleted until its retention period ends or its legal hold is removed.
Read the articleWhat is erasure coding?
Erasure coding protects stored data by splitting it into data and parity fragments on different servers, so it can be rebuilt after some fragments are lost.
Read the articleWhat is S3-compatible storage?
S3-compatible storage is object storage that accepts the S3 API, so existing S3 tools, SDKs and backup software work after you change the endpoint and keys.
Read the articleKubernetes
What is the Kubernetes Gateway API?
The Kubernetes Gateway API is a set of routing resources for getting traffic into a cluster, built as the successor to Ingress with roles and typed routes.
Read the articleWhat is a Kubernetes LoadBalancer Service?
A Kubernetes LoadBalancer Service exposes pods to clients outside the cluster through an external load balancer that the cloud provider creates and connects for you.
Read the articleWhat is an ingress controller?
An ingress controller is software in a Kubernetes cluster that reads Ingress objects and routes incoming HTTP and HTTPS traffic to the right Services.
Read the articleWhat is Kubernetes autoscaling?
Kubernetes autoscaling changes pod counts, pod resources or node counts automatically as load changes, so workloads keep up with peaks without paying for idle capacity.
Read the articleWhat is a pod disruption budget?
A pod disruption budget limits how many pods of a replicated Kubernetes app can be evicted at once, so node drains and upgrades keep it serving.
Read the articleWhat is Kubernetes version support?
Kubernetes version support is the period in which a minor release still gets patches and security fixes, about 14 months for each release.
Read the articleWhat is a CSI driver?
A CSI driver is a plugin that connects Kubernetes to a storage system, so clusters can create, attach, mount and resize volumes without storage code in Kubernetes.
Read the articleWhat is a persistent volume?
A persistent volume is Kubernetes storage that outlives the Pods using it, so databases and queues keep their data across restarts and rescheduling.
Read the articleWhat is a CNI plugin?
A CNI plugin connects Kubernetes pods to the network: it creates each pod's interface, assigns its IP address and often enforces network policies.
Read the articleWhat is a Kubernetes cluster upgrade?
A Kubernetes cluster upgrade moves the control plane and then the worker nodes to a newer Kubernetes version, one minor release at a time.
Read the articleWhat is a node pool?
A node pool is a group of Kubernetes worker nodes with the same machine type and settings, managed as one unit for sizing, upgrades and workload placement.
Read the articleWhat is the Kubernetes control plane?
The Kubernetes control plane manages a cluster's state through the API server, etcd, the scheduler and the controller managers. Worker nodes run pods; the control plane decides what runs where.
Read the articleWhat is managed Kubernetes?
Managed Kubernetes is Kubernetes where a provider runs the control plane, etcd and upgrades, while you deploy and run your own workloads on the worker nodes.
Read the articleNetwork
What is a layer 4 load balancer?
A layer 4 load balancer spreads TCP and UDP connections across backend servers using IP addresses and ports, without reading the application data inside.
Read the articleWhat is IP transit?
IP transit is a paid service in which a larger network carries your traffic to and from the rest of the internet and announces your IP ranges.
Read the articleWhat is CNAME flattening?
CNAME flattening lets a zone apex like example.com point at another hostname, by having the DNS server answer with that hostname's A and AAAA addresses.
Read the articleWhat is the ACME DNS-01 challenge?
The ACME DNS-01 challenge proves you control a domain by publishing a TXT record, so a certificate authority can issue TLS certificates, including wildcards.
Read the articleWhat is a CAA record?
A CAA record is a DNS record that names which certificate authorities may issue TLS certificates for a domain, so other authorities refuse to issue.
Read the articleWhat is DNSSEC?
DNSSEC adds digital signatures to DNS records, so resolvers can check that an answer came from the zone owner and was not changed on the way.
Read the articleWhat is a DNS zone file?
A DNS zone file is a plain text file that lists every record of a DNS zone, in a standard format nameservers load and providers import and export.
Read the articleOperations
What are SLAs, SLOs and SLIs?
SLAs, SLOs and SLIs describe service reliability: an SLI measures it, an SLO sets a target for that measure, and an SLA puts a target into a contract.
Read the articleWhat are RPO and RTO?
RPO and RTO are recovery targets: RPO caps how much data you can lose in an outage, RTO caps how long recovery may take.
Read the articleWhat is a failure domain?
A failure domain is a group of hardware that can fail together, such as a disk, server, rack or site, so copies of data must sit in separate ones.
Read the articleWhat is vendor lock-in?
Vendor lock-in is the cost and effort that keeps you on one cloud provider, built up through proprietary APIs, provider-specific identity and the price of moving data out.
Read the articleWhat are egress fees?
Egress fees are per-gigabyte charges for data leaving a cloud provider's network, usually to the internet, while data coming in is typically free.
Read the articleWhat is a cloud exit?
A cloud exit is the planned move of workloads and data off a cloud provider, onto another platform or your own servers, without losing data or service.
Read the article