ACA CCE flashcards
ACA CCE · Build recall with a focused review session.
32 cards. Read the term and the definition here. The deck at the top of this page is optional.
- Shared responsibility model
- A division of security duties: the cloud provider secures physical facilities, hardware, and virtualization, while the customer secures guest OS, identities, data, and applications.
- Elasticity
- The automatic addition and removal of capacity in response to real-time demand, so resources match load without manual provisioning or persistent over-provisioning.
- Availability zone
- A physically separate datacenter group within a region with independent power and network; deploying across zones improves fault tolerance without cross-region latency.
- IaaS
- Infrastructure as a Service delivers virtual machines, storage, and networking; the customer manages the operating system and everything above it.
- Horizontal scaling
- Adding more instances behind a load balancer to handle increased load, preferred for stateless tiers because it avoids single-node size limits and downtime.
- 99.99% availability
- An uptime target permitting roughly 52.6 minutes of downtime per year, commonly used to compare service-level commitments across cloud offerings.
- ECS
- Alibaba Cloud Elastic Compute Service, which provides resizable virtual machines offered in general-purpose, compute-optimized, memory-optimized, and storage-optimized families.
- OSS storage classes
- Object Storage Service tiers ranging from Standard through Infrequent Access, Archive, and Cold Archive, trading higher retrieval latency and cost for lower storage price.
- VPC and vSwitch
- A Virtual Private Cloud is an isolated network defined by a CIDR block; vSwitches are subnets within it, each bound to a single zone.
- Security group
- A stateful instance-level firewall for ECS that permits traffic by protocol, port range, and source; inbound rules default to deny.
- Server Load Balancer
- A managed traffic distributor across backend servers; CLB is classic layer 4/7, ALB targets HTTP/HTTPS applications, and NLB handles high-performance layer 4.
- Auto Scaling
- A service that adjusts ECS instance count in a scaling group based on scheduled or metric triggers such as CPU utilization thresholds.
- ApsaraDB RDS
- A managed relational database service that automates backups, patching, and high availability for engines such as MySQL, SQL Server, and PostgreSQL.
- RAM
- Resource Access Management, which controls which identities can perform which API actions on which resources through users, groups, roles, and policies.
- CDN
- A content delivery network caching static assets at edge nodes near users, reducing origin load and latency; it is commonly paired with OSS as origin.
- Model Studio
- Alibaba Cloud's platform for invoking foundation models such as the Qwen family through APIs, with model choice based on capability, context length, and price.
- Token-based billing
- Model API charges metered separately on input tokens and output tokens, with output tokens typically priced higher per token than input tokens.
- Context window
- The maximum number of tokens a model can process in one request, covering prompt plus completion; exceeding it requires chunking or summarization.
- Retrieval-Augmented Generation
- A pattern that retrieves relevant documents from a knowledge base and supplies them to the model so answers are grounded in current or private content.
- Embeddings
- Numeric vector representations of text that place semantically similar content close together, enabling similarity search when indexed in a vector store.
- Temperature
- A generation parameter controlling randomness; lower values produce more deterministic, repeatable output and higher values produce more varied output.
- Qwen models
- Alibaba Cloud's foundation model family, offered in multiple sizes and modalities including text, vision, and audio, allowing cost-capability tradeoffs.
- Least privilege
- Granting an identity only the permissions required for its task, implemented in Alibaba Cloud by scoping RAM policies to specific actions and resources.
- RAM role with STS
- A mechanism issuing temporary credentials so services or federated users gain scoped access without storing long-term AccessKey secrets in code.
- WAF vs Anti-DDoS
- WAF filters application-layer HTTP attacks such as SQL injection and XSS; Anti-DDoS mitigates volumetric network-layer floods. They address different threat layers.
- ActionTrail
- An audit service recording Alibaba Cloud API calls and account activity, supporting compliance reporting and forensic investigation of changes.
- KMS
- Key Management Service, which creates and manages encryption keys used for data at rest in services such as OSS and RDS.
- Subscription billing
- A prepaid commitment to a resource for a term such as a month or year at a lower unit price than on-demand usage.
- Pay-as-you-go
- Postpaid billing based on actual consumption with no upfront commitment, suited to variable or short-lived workloads but with a higher unit price.
- Data transfer out
- Charges for internet egress from Alibaba Cloud; traffic over internal endpoints within a region is typically not billed as internet egress.
- Cost allocation tags
- Key-value labels attached to resources so billing reports can attribute spend to specific projects, teams, or environments for cost management.
- Support severity levels
- Ticket classifications where production-down incidents receive the fastest response SLA, while general questions receive slower responses under the same plan.