怎样通过谷歌 PCA 考试(一 索引)
#gcp #certification
这次主要来讲一些自己通过 PCA 的一些心得。
基础知识和准备
首先，推荐给给大家一些常见的好的资料，相信大家在其他地方也看到过这些整理的内容：
学习材料合集
https://medium.com/@sathishvj/on-passing-all-google-cloud-certifications-54b2cc1e428c
A collection of posts, videos, courses, qwiklabs, and other exam details for all exams:
https://github.com/sathishvj/awesome-gcp-certifications
选择合适的服务
✔ You should be well-versed in understanding when to use which gcp service: Here are some of the important blogs/videos that you should have a look at:
- Which VPN service to choose, adequately explained by
Image by
- Choose the right load balancer as per your requirements
Image by
✔ In case of any IAM-related questions, you should always consider the principle of least privilege access while answering it.
✔ You should also consider cost-effectiveness and Google-recommended best practices while answering.
✔ You should always go with Cloud-native services as opposed to the open-source or third party unless the GCP service doesn’t satisfy the requirements. e.g In the case of Infrastructure as a Code service, you should select Deployment Manager instead of Terraform.
考点
在大家对谷歌云有了一定了解后，在下面着重讲解一些考点：
网络架构选择
- 负载均衡
- 网络互联:
- 跨 VPC 互联:
- 与自建机房互联: Dedicated Interconnect
- 与自建机房IP重叠时互联
计算资源选择
- GKE
- Cloud Function: 可以缩减到 0
- App engine: Standard 和 Flex 区别，如果需要连接到 on-prime，必须使用flex配合Cloud VPN
存储/数据库选择
- Storage: 用来存储长期的数据，转化为 Coldline 等类型降低成本。
- Regional Disk:
- Firebase: 无托管，使用 batch 操作，避免scan
- BigTable:
- 时序数据库优选
- 热点问题解决方法: 打散row_key
- Windows SQL server 高可用方案
数据处理
Dataflow: 实时数据处理
BigQuery: OLAP方案，支持SQL，适合在TB以上数量级进行数据分析
DataProc: 适合在已有数据处理，不需要做大改动即可迁移到GCP，使用 preemition vm 降低成本
其他
- IAM: 用户的权限是各个权限分配的并集(union)
- 账号迁移:
- Google Cloud Directory Sync 同步本地AD到谷歌云
- 迁移到GCP云的计划
练习题
最后，附上我在准备考试时练习过的题库链接：
- Case Study: ./how-pass-google-pca-2-case-study
- Question: