CPSC 4820 - DAY 9
FEBRUARY 22, 2018

================================================================================

Avoid single points of failure.

Should be able to handle switching to secondary devices in case of failure.

Recovery Time Objective - how much the system recover?

Recovery Point Objective - how much data can you afford to lose.

Fault tolerance - build int redundancy of an application's components.

Recoverability - The process, policies, and procedures related to restoring 
service after a catastriphic event.

Scalability - the ability of an application to accomodate growth without 
changing design.

In traditional, on-premesis IT, high availability:
	-very expensive
	-is suitable only for absolutely mission-critical applications.

AWS expands availability and recovery options by enabling you to use
	-Multiple servers
	-Isolated redundant data centers within Availability Zone
	-Multiple Availability Zones within each region
	-Regions across the globe.
	-Fault-tolerant services.

Default to use a SINGLE region unless a multi-region deployment is necessary.

If you use a single region, multi AZ is the bare minimum HA solution.

ELASTIC LOAD BALANCING
----------------------

	A managed load balancing service that distributes incoming application
traffic across multiple AWS EC2 instances.

	Elastic Load balancing distributes load between instances.

	Can be public or internal facing and are able to use HTTP, HTTPS, TCP

	Elastic ip addresses are designed for dynamic cloud computing and can
	be attached to amazon ec2 instances

	Enable you to mask the failure of an instance by allowing your users to
	use same IP address with replacement resources.

AMAZON ROUTE 53
---------------

	An authoritative DNS from AWS.

	Routing:
		-Simple Round robin - single server environments.
		-Weighted round robin - assign weights to resource record set
		to specify the frequency.
		-Latency-based routing helps to improve your global application