Alibaba ACA Cloud Operator flashcards
Alibaba ACA Cloud Operator · Build recall with a focused review session.
33 cards. Read the term and the definition here. The deck at the top of this page is optional.
- Shared Responsibility Model
- Alibaba Cloud secures physical facilities, hardware, and virtualization; customers secure guest operating systems, applications, data, identities, and network configuration within their VPCs.
- Region and Zone
- A Region is a geographic area containing multiple isolated Zones; each Zone has independent power and network, so multi-Zone designs survive single-datacenter failures.
- Resource Group
- A logical container for organizing resources so that RAM permissions and cost reports can be scoped to a group rather than to the whole account.
- AccessKey Pair
- An AccessKey ID plus Secret used to sign Alibaba Cloud API requests; long-lived pairs should be replaced by temporary STS tokens for RAM users.
- Idempotent Operation
- An operation that produces the same end state when repeated. ROS stacks and API retries depend on idempotency so partial failures can be safely re-applied.
- Change Rollback Path
- A documented method to restore the previous state after a failed production change, such as reverting a stack or restoring an ECS image snapshot.
- ECS Instance Family
- A category of instance types sharing a vCPU-to-memory ratio: g-series general purpose, c-series compute optimized, r-series memory optimized.
- Subscription Billing
- A prepaid ECS billing mode where capacity is committed for a fixed term, typically yielding the lowest unit price but requiring upfront payment.
- Pay-As-You-Go Billing
- A postpaid ECS billing mode charged by usage with no long-term commitment, offering flexibility at a higher unit price than subscription.
- Spot Instance
- A deeply discounted ECS instance that can be reclaimed by the platform with short notice, suitable for fault-tolerant batch or stateless workloads.
- Auto Scaling Group
- A group of ECS instances managed together, scaling in or out according to CloudMonitor metric alarms or scheduled policies and bound to a launch template.
- Custom Image
- A Region-scoped template capturing an ECS instance's system and data disks, used to launch identical instances or to restore a prior state.
- Security Group
- A stateful virtual firewall attached to ECS instances whose inbound rules default to deny and match on protocol, port range, and source.
- Elastic IP Address
- A public IP that can be detached from one resource and re-bound to another, such as an ECS instance or NAT Gateway, within the same Region.
- Stop Versus Release
- Stopping a Pay-As-You-Go instance keeps its disks and may still incur storage charges; releasing destroys the instance and non-retained disks permanently.
- OSS Storage Classes
- Standard, Infrequent Access, Archive, and Cold Archive tiers that trade lower storage cost for higher retrieval latency and longer minimum storage durations.
- OSS Bucket Naming
- Bucket names must be globally unique across all Alibaba Cloud accounts and are tied to a Region that determines data residency and endpoint.
- ESSD Performance Levels
- Enhanced SSD disks offer PL0 through PL3 performance levels with increasing IOPS and throughput; AutoPL scales performance with capacity and burst.
- Snapshot Incrementality
- ECS snapshots capture only changed blocks after the first full snapshot; they persist independently, so deleting the source disk does not delete its snapshots.
- VPC and VSwitch
- A VPC is an isolated network with its own CIDR block; each VSwitch maps to one Zone and holds instances that communicate subject to routes and security groups.
- NAT Gateway
- A VPC component that lets private instances without public IPs initiate outbound internet traffic while blocking unsolicited inbound connections.
- Server Load Balancer
- A service distributing traffic to backend servers through L4 or L7 listeners, using health checks to automatically remove unhealthy backends from rotation.
- Route Table Precedence
- Each VSwitch uses one route table; when multiple routes match, the most specific prefix wins, and custom routes can direct traffic to NAT or transit devices.
- CloudMonitor Metric Granularity
- CloudMonitor gathers ECS metrics at one-minute intervals; hypervisor-level metrics such as CPU are automatic, while guest memory and processes need the agent.
- Alarm Rule Components
- An alarm rule pairs a metric with a statistic, comparison operator, threshold, consecutive evaluation periods, and a notification contact group.
- INSUFFICIENT_DATA State
- An alarm state reported when a metric has no datapoints in the evaluation window, distinguishing missing telemetry from a healthy OK state.
- CloudMonitor Agent
- Software installed inside the guest OS that reports memory, process, and other operating-system metrics that the hypervisor cannot observe directly.
- Log Service Query
- A Log Service query combines a full-text search statement with an optional SQL analytics clause evaluated over a user-specified time range.
- Event Monitoring
- CloudMonitor system events record state changes such as instance restarts or disk exhaustion, complementing threshold-based metric alarms.
- RAM Policy Elements
- A RAM policy specifies Effect, Action, Resource, and optional Condition; an explicit Deny always overrides any Allow statement in the evaluation.
- Instance RAM Role
- A role attached to an ECS instance that grants temporary credentials to applications, avoiding embedded long-lived AccessKeys in code or configuration.
- ActionTrail Audit
- A service recording API-level operations with caller identity, timestamp, and source IP, deliverable to OSS or SLS for compliance retention.
- Cost Allocation Tags
- Consistent key-value tags such as env or owner that let Cost Management split spend by team or workload and drive tag-based automation.