Introductory Flashcards
What document defines common charateristics of cloud computing
Special Publication 800-145
Essential chracteristics of cloud momputing according to NIST Model
On demmand self-service
Broad network access
Resource pooling
Rapid elasticity
MEasured Service
On demand self Service
Customer can provision capability themselves
Broad network access
Capabilities avaliable through standard mechanisms
Hetregrenous client platforms
Resource pooling
*Computing resource pooled to server multiple customers
*No control of exact location
*Can define some high levels of asbstraction (storage, memroy etc)
Rapid elasticity
Capabilities can be provisioned and released flexibly and quickly
What NIST charateristics gives the impression of unlimited resources
Rapid elasticity
Measured service
Leveraging a meter service
Billed for what you use
Why do we need cloud?
- Too much data
- Too much compute
- Lack of scalability
Benefits of AWS
Enterprise friendly
service
Easy access to
resources
Increase in speed and
agility
Benefits of Azure
Better development
operations
Strong Security profile
Provides a costeffective solution
Benefits of GCP
*Better pricing than
competitions
* Better advanced machine learning / ai / big data stuff
* Redundant backups
Disadvantages of Azure
Different codebase for
cloud and premise
PaaS ecosystem is
not efficient as IaaS
Poor management of
GUI and tools
Disadvantages of GCP
Support fee is quite
expensive
Complex pricing
schema
Downloading data
from Google Cloud
Storage is expensive
refers to the execution of more than one procedure at the same time
Concurrency
more than one CPU or
CPU core, and makes progress on more than one task simultaneously
Parallel execution
What type of execution is show in image
Parallel Concurrent Execution
application splits its tasks up into smaller subtasks which can be processed in parallel
Parallelism
Advantages of parallelism
Saves time
Solve larger problems
Disadvantages of parallelism
Some data incompatible
Costly in terms of hardware
COmplex
A large task is broken into
several discrete tasks
Each task is computed in the
individual computer of a
distributed system
Distributed COmputing
Disadvantages of distributed computing
Infra costs
Security issues
Increased troubleshooting
Increased support complexity
Compuete Engine VM an example of what cloud model
IaaS
Google App Engine an example of what cloud model
PaaS
The following are benfits of what Google service:
Auto Scaling – No need to over provision
Easy Logs- View logs in web console
Easy Deployment- Literally 1-click deploy
Free Quota- 99% of app will pay nothing
Affordable Scaling-Price better than AWS
No config- No need to config OS or servers
Easy Security – Google patches OS/servers
Google App Engine
Explain App Engine Architecture
App Engine API
Talk to Datastore, Memcache, Cloud Storage
Types of app engine enviroments
Standard
Flexible
Advantages of on-prem deployment
- CAPEX
- Complete control
- Do not rely on itnernet connectivity
Shortages of on-prem deployment
- Large CAPEX
- Reponsibility
- Longer implementation
IaaS use cases
- Legacy apps
- Big Data
- Disaster REcovery
- Network Services
IaaS Advantages
- Fast access to infra
- Flexiblity
- Portability
IaaS Disadvantage
- More ops/infra needed
- Not as optimised as PaaS
- Vendor outages
PaaS advantages
- Simplicity
- Scalable
- Easy can customize apps without much overhead
PaaS disadvantages
- Vendor lock-in
- Lack of control
- hardwardware and softwre dependes
SaaS Disadvantages
- Security Implications
- Limited customisation
- Preferences
Cloud Deployment Models
- Public Cloud
- Private Cloud
- Community Cloud
- Hybrid Cloud
Advantages of public clouds
- Cost Effective
- No maintinace
- Near-unlimited scalability
- Reliable
Advantages of private clouds
- More flexibility
- More control
- Provide enchanced level of security and privacy
Disadvantages of private cloud
- Responsibility for own infrastructure
- Cost
The cloud infrastructure is shared by several organizations that have shared concerns
Community CLoud
Disadvantages of Hybrid Cloud
- Difficult to implement and maintain
- Complexity
- Visiblity
- Hardware expenses - duality of both pub and priv
Hadoop enables distributed across
Hundreds of thousands of servers
Who developed Hadoop
Apache
Hadoop enables the following characteristics
Scalability
Flexibility
SPeed
How is data stored (input/output) in Hadoop?
Hadoop Distributed File System (HDFS)
Hadoop - node that manages the Hadoop distirbuted file system
NamedNode
Hadoop - Job
Program executed on Hadoop across a dataset
In Hadoop what is provided by the user?
Job COnfiguration
In Hadoop what does the Job COnfiguration define?
Input FOrmat
Map functions
Number of reduce tasks
Reduce functions
Output type
