- EC2 elastic compute cloud
- Mnemonic - fight dr mc px z au
- FPGA - Genomic Research, Financial analysis
- IOPS - Database/ Applications
- Graphics - 3D modeling
- High disk throughput
- t2.micro - general purpose - free tier
- density
- ram
- m main general purpose
- c compute intensive more CPU
- p graphics
- x Extreme Memory
- z Extreme Memory and Extreme CPU
- A ARM based architecture
- U Bare Metal Servers
- IAM - User Group Role, Policies
- Virtual Private Cloud (VPC) 5 per account, public(route with IG) private subnet(default route), NAT gateway 10gb-HA-AWS managed / NAT instance (AMI), NACL(stateless inbound/outbound need to be specifically mentioned), Security Group(state full), site to site VPN, direct connect via dc provider (1 or 2 per region)
- EBS general purpose(10K IOPS(3 IOPS / GB)) bootable, provisioned IOPS (>10K IOPS), HDD Throughput intensive, HDD capacity oriented, magnetic HDD (bootable)
- Logical to physical mapping of AZ is different for different account
- SQS first service, Message based, 14 days maximum retention, Pull. SWF(simple workflow service) Task based, 1yr maximum retention. SNS push. SES for only email notification.
- CloudWatch (performance), CloudTrail(API calls log), config(consistency in the configuration) logs for CloudTrail go to S3 where Athena can be used to retrieve logs using SQL. CloudWatch logs also go to S3 and they are retained indefinitely.
- System manager for cloud and on premise
- Migrations Tools to AWS
- Server migration service (SMS OVF)instance on VMware or Hyper-V block migration from on premise to cloud.
- Database migration service(DMS) instance on ec2 migration from on premise DB to cloud.
- Storage gateway as instance on premise as VM, that allows on premise to use cloud storage as NFS/SMB(S3 - Storage gateway configured for file), iSCSI (EBS - Storage gateway configured for block) VTL(S3/Glacier - Storage gateway configured for VTL).
- Snowball 50tb/80tb. Snowball Edge 100TB with compute power. Snowmobile exabyte of data
- CI/CD is methodology that allows developers to store code in a repository and collaborate with others. Task of compiling code and deploying application is fully automated and each task is orchestrated. In AWS this is analogous to Code Pipeline that consists CodeCommit(based on git) is repository for code, maintain different versions code and collaborate with other programmer, CodeBuild(kicks in as soon as the code commit happened), CodeDeploy(after the application is built it deployed either in rolling upgrade or Blue/Green - In Blue green Old Application and new applications are run in parallel with more weight given to new application, once new application tested in the field, old application is removed. All process starting from code commit to code build, and deploying application is fully automated.
- NoSQL - key value good for large number of data, no need for normalization 1nf,2nf,3nf. RDBM consistency model is ACID(Atomic, Consistency, Isolation, Durable) and NoSQL consistency model is BASE (Basic availability, Soft state and Eventual consistence).
Bye...
No comments:
Post a Comment