Horizontal Dynamic Scalability
1. Automated Scaling Listener -monitors capacity.
2. Listener takes decision based on pre-defined scaling policy.
3. Listener replicates resources.
ELASTIC RESOURCE CAPACITY ARCHITECTURE
Dynamic provisioning of virtual servers
- Fluctuating processing requirements
- Allocates and reclaims CPUs, RAMs
- Run-time of virtual server is monitored
- Additional resources leveraged before capacity threshold reached
- Virtual server and It resources vertically scaled
- Use Intelligent Automation Engine - sends scaling request to hypervisor or VIM
Main Components of Elastic Resource capacity arvhitecture.
SERVICE LOAD BALANCING ARCHITECTURE (CLOUD PROVIDERS)
A specialized variation of the workload distribution architecture for scaling cloud service implementations
Redundant cloud service deployments are created with a load balancer to dynamically distributes workloads.
2 types of load balancer:
- Independent of the cloud services and their host servers
- Built-in as part of application or server’s environment ( a primary server serves as a load balancer and communicates with neighboring servers to balance workload)
Cloud Bursting Architecture
Ø To meet higher usage demand, dynamic scaling or "bursts out" of on-premise resources to cloud based resources
Ø Cloud services are redundant, pre-deployed and inactive until cloud bursting
Ø Releases or "bursts in" to on-premise when no longer required(the reverse)
Ø Based on:
- Automated scaling listener to redirect requests to cloud-based services
- Resource replication to maintain synchronicity between on-premise and cloud-based resources.
Elastic Disk Provisioning Architecture
- Typically, cloud consumers are charged based on fixed disk storage allocation
- Aim - achieve dynamic storage provisioning and charge based on actual usage