Exoscale Certified Cloud Professional key terms

A managed group of identical Compute instances that can scale in or out automatically and can be spread across zones, so you manage capacity as a unit instead of individual machines.

A stored definition of instance settings such as template size, disk, network and startup script, used as the recipe when an instance pool creates new members.

Text supplied when an instance boots that cloud-init executes or applies, typically to install packages, set users or configure services without a manual login.

A rule that keeps members of a group on separate physical hosts, reducing the chance that one hardware failure takes down several of your instances at once.

A registered image of an operating system or preconfigured disk that belongs to your account and can be used to launch instances reproducibly.

A public IPv4 address reserved to your account that you can attach to or detach from resources, keeping the address stable across instance rebuilds and replacements.

A stateful set of ingress and egress rules attached to a resource that permits or blocks traffic by protocol, port and source or destination.

An isolated layer-2 network segment you create inside a zone, to which instances attach and receive private addresses so they can talk without traversing the public internet.

An independent failure and location domain within a region, with its own power and network, used to place resources for latency and resilience decisions.

A geographic area containing one or more zones where services are offered, and the unit that determines where data is stored and processed.

A network-attached disk you can create, snapshot and attach to a single instance, decoupling persistent data from the life of the compute node.

A point-in-time copy of a block storage volume stored independently, used to restore data or to create new volumes.

A named container in the S3-compatible object store that holds objects and carries its own access policy and visibility settings.

A time-limited URL generated with credentials that grants temporary access to a specific object without revealing the underlying keys.

A JSON document attached to a bucket that grants or denies actions to principals, controlling access independently of the credentials a client holds.

A service model in which the provider operates the database engine, backups, patching and failover while the customer manages schemas, users and queries.

A service tier that fixes the engine, compute size, storage and redundancy of a database, so scaling means migrating to a different plan.

A stored copy of a managed database used to restore it to a prior point in time or to a new service, retained according to the plan's policy.

A managed intermediary that reuses a limited set of database connections across many clients, protecting the database from connection exhaustion.

A named set of permissions that a principal assumes temporarily to obtain credentials, avoiding long-lived keys embedded in applications.

A credential pair, typically a key and secret, used to authenticate programmatic requests to the platform's API and CLI.

A login control requiring a second proof of identity beyond a password, such as a one-time code, before account access is granted.

A chronological record of actions taken in an account, showing who did what and when, used for investigation and compliance evidence.

The platform's managed AI offering, combining compute resources with model serving and tooling so teams can run inference and training without assembling infrastructure themselves.

A compute instance with attached graphics processors, chosen for workloads such as model training, fine-tuning and high-throughput inference.

A served, addressable interface to a deployed model that applications call to obtain predictions, scaling independently of the client.

The requirement that data be stored and processed within a specified jurisdiction, satisfied by choosing the region where resources are created.

European data protection law governing processing of personal data, shaping contracts, retention and location choices for workloads serving EU residents.

A contract between customer and provider setting out how personal data is handled, including subprocessors, security measures and breach notification.

The division of duties in which the provider secures the underlying platform and the customer secures their configurations, identities and data.

Control over where data lives and who can access it, including protection from foreign legal orders, achieved through jurisdiction, ownership and operational independence.

A public page publishing certifications, audit reports and security practices so customers can assess compliance before committing to the platform.

An international standard for information security management systems, certifying that an organization runs documented, audited controls over its information assets.

Adjusting the number of running instances in a pool based on metrics or a schedule, so capacity tracks demand without manual intervention.

An approach where every platform capability is exposed through a documented API, with the portal and CLI built on the same interface.

An object storage interface that follows the widely used S3 protocol, letting standard SDKs and tools work without code changes.

A record mapping an IP address back to a hostname, often required by mail servers and useful for identifying your resources in logs.