Alibaba Cloud Architect Professional flashcards

32 cards. Read the term and the definition here. The deck at the top of this page is optional.

ECS general purpose family

The g-series instance family balances vCPU and memory at roughly 1:4, suiting web servers, app servers, and mid-size databases that need neither extreme compute nor extreme memory.

Application Load Balancer

ALB is a layer-7 load balancer supporting HTTP, HTTPS, and gRPC with advanced routing rules like path, host, and header-based forwarding, plus weighted canary traffic splitting.

Network Load Balancer

NLB operates at layer 4 with very high throughput and low latency, supports static IP addresses, and is suited for TCP/UDP workloads and extreme connection volumes.

Multi-AZ deployment

Distributing ECS and RDS instances across multiple availability zones within one region so a single zone failure does not take down the whole application.

Auto Scaling scaling group

A logical collection of ECS instances with shared configuration, min and max size, and scaling rules that add or remove instances based on schedules or metric alarms.

VPC NAT Gateway

A managed component that lets instances in private subnets initiate outbound internet traffic without public IPs, while blocking unsolicited inbound connections.

Function Compute

A serverless compute service that runs event-triggered code with sub-second billing and automatic scaling, best for bursty or infrequent workloads rather than long-running stateful services.

RDS read-only instance

A replica of a primary RDS instance used to offload read queries; it applies changes asynchronously, so it can lag behind the primary during heavy write loads.

Server Migration Center

SMC migrates physical, virtual, or cloud servers into Alibaba Cloud ECS by installing an agent, syncing disks, and launching a target image.

Data Transmission Service

DTS performs full plus incremental database migration and synchronization with minimal downtime, and also supports ongoing replication between source and target databases.

Blue-green deployment

Running two identical environments where traffic is switched from the old stack to the new stack via load balancer or DNS, enabling instant rollback by switching back.

Canary release

Gradually shifting a small percentage of traffic to a new version using weighted routing, monitoring error rates, then increasing weight until fully rolled out.

Resource Orchestration Service

ROS is an infrastructure-as-code service that provisions and manages groups of Alibaba Cloud resources from JSON or YAML templates in a repeatable, versioned way.

Express Connect

A dedicated private network connection between an on-premises data center and Alibaba Cloud, offering more consistent latency and bandwidth than internet-based VPN.

VPN Gateway

An IPsec-based encrypted tunnel over the public internet connecting on-premises networks to a VPC; cheaper than a dedicated line but dependent on internet quality.

Container Service for Kubernetes

ACK is a managed Kubernetes service that handles control plane operations, node scaling, and integration with Alibaba Cloud networking, storage, and load balancing.

OSS Data Online Migration

A managed service that migrates object data from other cloud storage or on-premises sources into OSS over the network, supporting incremental and scheduled sync.

CloudMonitor

The platform service that collects metrics from ECS, RDS, SLB, and other resources, evaluates alarm rules, and triggers notifications or Auto Scaling actions.

Log Service

SLS collects, stores, indexes, and queries logs from applications and cloud services, supporting real-time analysis and long-term retention for troubleshooting and audit.

OSS storage classes

Object storage tiers from Standard through Infrequent Access, Archive, Cold Archive, and Deep Cold Archive trade retrieval time and minimum storage duration for lower per-GB cost.

Reserved instance

A prepaid capacity commitment for steady-state ECS workloads that lowers the hourly cost compared with pay-as-you-go billing in exchange for a term commitment.

Spot instance

A discounted ECS instance that may be reclaimed by the provider with short notice, suitable only for fault-tolerant, checkpointed, or stateless batch workloads.

Recovery time objective

RTO is the maximum acceptable duration to restore a service after failure; it drives choices like multi-AZ versus multi-region and active-passive versus active-active designs.

Recovery point objective

RPO is the maximum acceptable amount of data loss measured in time; it determines backup frequency and whether synchronous or asynchronous replication is required.

Auto Scaling cooldown

A waiting period after a scaling activity during which further scaling actions are suppressed, preventing rapid oscillation while new instances finish booting and warming up.

RAM role

An identity assumed by trusted entities to obtain temporary STS credentials, used for cross-account access and service-to-service authorization without distributing long-term access keys.

Security group

A stateful, instance-level virtual firewall controlling inbound and outbound traffic by allow rules on IP, port, and protocol; return traffic is automatically permitted.

Network ACL

A stateless subnet-level firewall that evaluates allow and deny rules for traffic entering or leaving a subnet, complementing instance-level security groups.

Key Management Service

KMS manages encryption keys and provides envelope encryption APIs so services like OSS and RDS can encrypt data with customer-managed or service-managed keys.

ActionTrail

A service that records API calls made in an account, including caller identity, source IP, and time, supporting security investigation and compliance auditing.

Web Application Firewall

WAF inspects HTTP and HTTPS traffic at layer 7 to block SQL injection, cross-site scripting, and malicious bots, protecting applications behind SLB or ALB.

Cloud Security Center

A centralized security service that performs baseline configuration checks, vulnerability scanning, and anomaly detection across ECS instances, containers, and other resources.