Cv0-04 section 2

Virtualization

Using software to pretend one physical computer is actually multiple separate “virtual” computers. Like having imaginary friends who can each use the same toy box without mixing toys.

Virtual Machine (VM)

A complete computer simulated in software. It has its own operating system and apps, separate from the physical machine and other VMs

Container

A way to package an application and everything it needs (code, tools, libraries) so it can run reliably anywhere. Containers share the host computer OS, making them smaller and faster than VMs.

Hypervisor

The software that creates and manage VMs

Type 1 hypervisor

Runs directly on the computer hardware (like bare metal). More efficient, used in data centers.

Type 2 hypervisor

Runs as an application on top of an existing OS (like running VirtualBox on your windows laptop). Easier for personal use

Virtual networking

Creating networks using software instead of physical cables and switches. Allows creating virtual switches, routers, and isolated network segmented (like VLANs or VPCs) in the cloud.

VM Templates

A pre-made recipe for creating VMs with specific settings (OS, software, resources). Faster and more consistent than building each VM manually.

Auto-scaling

Automatically adding or removing computer resources (like VMs or containers) based on demand. Like automatically adding more cashiers when the store gets busy.

Provisioning

The process of setting up and allocating cloud resources (like VMs, storage, networks). Often automated using tools or code.

vCPU (virtual CPU)

The amount of processing power assigned to a VM from the physical computer’s CPU(s).

vGPU (Virtual GPU)

Assigning graphics processing power to VMs for tasks like 3D rendering or machine learning

Oversubscription

Assigning more total resources (like CPU or memory) to all the VMs on a host than the host physically has. Works okay if not all VMs need their maximum resources at the same time, but can cause slowness if they do.

Instances

A term often used for a single VM running in the cloud, usually with a specific combination of CPU, memory, storage, and networking.

Storage I/O (input/output)

Reading data from or writing data to storage

Performance baselines

Historical metrics to detect anomalies

IOPS (I/O Operations per second)

How many read/write actions a storage device can do per second. A key measure of storage speed, especially for databases

Right-sizing resources

Matching vCPU/memory/storage to actual workload

CPU ready time

How long a VM waits for CPU scheduling

Throughput

How much data can be transferred per second (like MB/sec). Important for large or streaming

Latency

The delay in data transfer; how long it takes for data to get from A to B. Lower is better. Affected by distance and network congestion.

Bucket

A common term (especially in object storage like AWS S3) for a container that holds your data. Buckets usually have globally unique names

Full clone

A full clone is a completely separate copy of a VM that shares no system resources with the parent once it's running.

Linked clone

A linked clone, on the other hand, continues to share virtual disks with the parent after it's created. Since it runs independently, a full clone generally has faster performance than a linked one.

Snapshots vs backup

Point-in-time image vs full data copy

vApps

Grouping multiple VMs with shared resource controls

VM Affinity rules

Keep certain VMs together or apart on hosts

Nested virtualization

Running a hypervisor inside of a VM