CS CLOUD COMPUTING EXAN 1

Cloud Computing

On-demand delivery of computing services over the internet with pay-as-you-go pricing.

Two Essential Characteristics

(1) On-demand self-service, (2) Broad network access.

IaaS

Provides compute, storage, and networking; customer manages OS & applications.

PaaS

Offers a development environment; provider manages infrastructure & OS.

Public Cloud

Shared resources, cost-effective, scalable (e.g., AWS, Azure).

Private Cloud

Dedicated to one organization, more secure but costly.

Amazon EC2

Introduced pay-as-you-go virtual server instances, boosting cloud adoption.

Hypervisor

Software that manages virtual machines by abstracting hardware.

Two Advantages of Containers

Faster startup (no separate OS) and Efficient resource usage (shared kernel).

One Drawback of Containers

Weaker isolation than VMs due to shared host OS.

Ideal Container Use Case

Microservices-based apps for rapid deployment & scaling.

Horizontal Scaling

Adding more instances.

Vertical Scaling

Upgrading machine resources.

Load Balancer

Distributes traffic across servers to improve reliability & performance.

Auto-Scaling

Adjusts active resources based on real-time demand for stability.

Shared Responsibility Model

Cloud Provider: Security of physical infrastructure & platform. Customer: Secure configurations, access management, data protection.

Cloud Storage

Object Storage: Stores data as unique-key objects (e.g., Amazon S3). Block Storage: Raw storage volumes like virtual disks (e.g., Amazon EBS).

Two Common Cloud Threats

Unauthorized Access - Weak credentials, brute-force attacks. Data Breaches - Misconfigurations, poor encryption.

Security Mitigation Strategies

Use MFA & role-based access control. Enable encryption & monitoring tools.

Hybrid Cloud Strategy

Keep sensitive data in private cloud. Use public cloud for scalable workloads.

Hybrid Cloud Challenges

Complex integration between clouds. Consistent security monitoring.

Two Pricing Models

On-Demand: Pay for usage (good for unpredictable workloads). Reserved Instances: Commit for 1-3 years for cost savings.

Reserved Instances

Lower cost for predictable workloads due to long-term commitment.

Spot Instances

Cheaper but can be interrupted; ideal for batch processing.

Cloud Cost Optimization

Right-Sizing: Adjust resource allocation. Monitoring Tools: Track usage & set alerts.

Monitoring Tools: Track usage & set alerts.

Cloud Migration Lesson

Re-architecting is crucial; 'lift & shift' without planning can be inefficient.