UNIT - 3

Cloud resource provisioning and SLA (Service Level Agreement) management are crucial
aspects of cloud computing. Here’s a brief overview of each:
Cloud Resource Provisioning
Cloud resource provisioning involves allocating and configuring IT resources to meet the
dynamic needs of an organization. This process ensures that resources such as CPU, memory,
storage, and network bandwidth are available as needed. There are several types of
provisioning:

1. Static Provisioning:

 Resources are allocated in advance based on expected demand. This is suitable for
applications with predictable workloads.
 Static provisioning can be used successfully for applications and typically constant
demands or workloads.
 The cloud provider allows the customer with a set number of resources. The client
can thereafter utilize these resources as required. The client is in charge of making
sure the resources aren’t over utilized.
 This is an excellent choice for applications with stable and predictable needs or
workloads.
 Either a one-time cost or a monthly fee is applied to the client.

2. Dynamic Provisioning:

 The provider adds resources as needed and subtracts them as they are no longer
required.
 It follows a pay-per-use model, i.e. the clients are billed only for the exact resources
they use.
 Resources are allocated on-demand, allowing for flexibility and scalability.
 Consumers must pay for each use of the resources that the cloud service provider
allots to them as needed and when necessary.
 This is ideal for applications with variable workloads.
 This is a suitable choice for programs with erratic(unpredictable) and shifting
demands or workloads.

3. Self-service provisioning or user self-provisioning: In user self-provisioning,

sometimes referred to as cloud self-service, the customer uses a web form to acquire
resources from the cloud provider, sets up a customer account, and pays with a
credit card. Shortly after, resources are made accessible for consumer use.

Importance of Cloud Provisioning:

 Scalability: Being able to actively scale up and down with flux(continuous change)
in demand for resources is one of the major points of cloud computing
  Speed: Users can quickly spin up multiple machines as per their usage without the
need for an IT Administrator
 Savings: Pay as you go model allows for enormous cost savings for users, it is
facilitated by provisioning or removing resources according to the demand
SLA Management
SLAs are formal agreements between service providers and customers that define the
expected level of service. Effective SLA management ensures that cloud services meet these
agreed-upon standards. Key components include:

1. Performance Metrics: These include uptime, response time, and throughput, which are
monitored to ensure compliance with the SLA.
2. Penalties and Remedies: These are predefined actions that occur if the service provider fails
to meet the SLA terms.
3. Monitoring and Reporting: Continuous monitoring of services and regular reporting help in
maintaining transparency and trust.

Challenges of Cloud Provisioning:

 Complex management: Cloud providers have to use various different tools and
techniques to actively monitor the usage of resources
 Policy enforcement: Organisations have to ensure that users are not able to access the
resources they shouldn’t.
 Cost: Due to automated provisioning costs may go very high if attention isn’t paid to
placing proper checks in place. Alerts about reaching the cost threshold are required.
 Challenges and Solutions

 Resource Over-Provisioning: Allocating more resources than needed can lead to increased
costs. Dynamic provisioning helps mitigate this by adjusting resources based on real-time
demand.
 SLA Violations: Ensuring compliance with SLAs can be challenging. Implementing robust
monitoring tools and automated management systems can help in early detection and
resolution of potential issues

Tools for Cloud Provisioning:

 Google Cloud Deployment Manager

 IBM Cloud Orchestrator
 AWS Cloud Formation
 Microsoft Azure Resource Manager
Service level agreements in Cloud computing
A Service Level Agreement (SLA) is the bond for performance negotiated between the
cloud services provider and the client. Earlier, in cloud computing all Service Level
Agreements were negotiated between a client and the service consumer. Nowadays, with the
initiation of large utility-like cloud computing providers, most Service Level Agreements are
standardized until a client becomes a large consumer of cloud services. Service level
agreements are also defined at different levels which are mentioned below:
 Customer-based SLA
 Service-based SLA
 Multilevel SLA

 Customer-based SLA: A business agreement with an individual client or group that

sets a specific standard for support. For example, a telecommunications company
might put a clause in its biggest customer’s contract that guarantees a specific uptime.

 Service-based SLA: A specific standard for support that’s promised to all customers
who are using a particular service or product. If the telecom company guarantees a
specific uptime to all clients or offers a free routine service to every customer, that
would be a service-based SLA.

 Multilevel SLA: A multipart agreement that’s split into different levels, each
addressing different sets of customers. A multilevel SLA might have a corporate-level
component that applies to all users, as well as a separate agreement that covers a
specific customer group. The telecom company may provide basic services and
support to all its customers, but it could also offer different price ranges that dictate
different service levels.

Multilevel SLA: The SLA is split into the different levels, each addressing different set
of customers for the same services, in the same SLA.

 Corporate-level SLA: Covering all the generic service level management (often
abbreviated as SLM) issues appropriate to every customer throughout the
organization. These issues are likely to be less volatile and so updates (SLA reviews)
are less frequently required.
 Customer-level SLA: covering all SLM issues relevant to the particular customer
group, regardless of the services being used.
 Service-level SLA: covering all SLM issue relevant to the specific services, in
relation to this specific customer group.
Service Level Agreements usually specify some parameters which are mentioned below:
1. Availability of the Service (uptime)
2. Latency or the response time
3. Service components reliability
4. Each party accountability
5. Warranties
In any case, if a cloud service provider fails to meet the stated targets of minimums then the
provider has to pay the penalty to the cloud service consumer as per the agreement. So,
Service Level Agreements are like insurance policies in which the corporation has to pay as
per the agreements if any casualty occurs.

SLA Lifecycle
Steps in SLA Lifecycle

1. Discover service provider: This step involves identifying a service provider that can
meet the needs of the organization and has the capability to provide the required service.
This can be done through research, requesting proposals, or reaching out to vendors.
2. Define SLA: In this step, the service level requirements are defined and agreed upon
between the service provider and the organization. This includes defining the service
level objectives, metrics, and targets that will be used to measure the performance of the
service provider.
3. Establish Agreement: After the service level requirements have been defined, an
agreement is established between the organization and the service provider outlining the
terms and conditions of the service. This agreement should include the SLA, any
penalties for non-compliance, and the process for monitoring and reporting on the service
level objectives.
4. Monitor SLA violation: This step involves regularly monitoring the service level
objectives to ensure that the service provider is meeting their commitments. If any
violations are identified, they should be reported and addressed in a timely manner.
5. Terminate SLA: If the service provider is unable to meet the service level objectives, or
if the organization is not satisfied with the service provided, the SLA can be terminated.
This can be done through mutual agreement or through the enforcement of penalties for
non-compliance.
6. Enforce penalties for SLA Violation: If the service provider is found to be in violation
of the SLA, penalties can be imposed as outlined in the agreement. These penalties can
include financial penalties, reduced service level objectives, or termination of the
agreement.

Advantages of SLA
 1. Improved communication: A better framework for communication between the service
provider and the client is established through SLAs, which explicitly outline the degree of
service that a customer may anticipate. This can make sure that everyone is talking about
the same things when it comes to service expectations.
2. Increased accountability: SLAs give customers a way to hold service providers
accountable if their services fall short of the agreed-upon standard. They also hold service
providers responsible for delivering a specific level of service.
3. Better alignment with business goals: SLAs make sure that the service being given is in
line with the goals of the client by laying down the performance goals and service level
requirements that the service provider must satisfy.
4. Reduced downtime: SLAs can help to limit the effects of service disruptions by creating
explicit protocols for issue management and resolution.
5. Better cost management: By specifying the level of service that the customer can
anticipate and providing a way to track and evaluate performance, SLAs can help to limit
costs. Making sure the consumer is getting the best value for their money can be made
easier by doing this.

Disadvantages of SLA

1. Complexity: SLAs can be complex to create and maintain, and may require significant
resources to implement and enforce.
2. Rigidity: SLAs can be rigid and may not be flexible enough to accommodate changing
business needs or service requirements.
3. Limited service options: SLAs can limit the service options available to the customer, as
the service provider may only be able to offer the specific services outlined in the
agreement.
4. Misaligned incentives: SLAs may misalign incentives between the service provider and
the customer, as the provider may focus on meeting the agreed-upon service levels rather
than on providing the best service possible.
5. Limited liability: SLAs are not legal binding contracts and often limited the liability of
the service provider in case of service failure.
Cloud Billing overview

Cloud Billing is a collection of tools that help you track and understand your Google Cloud
spending, pay your bill, and optimize your costs.

Cloud billing is a method of generating bills from resource usage data in a cloud
environment. This approach to billing allows for automated, scalable and flexible
management of billing operations. It's especially useful for services such as software,
infrastructure and online platforms. It's a dynamic, adaptable and efficient way for
businesses to handle their billing needs, especially in an environment where services and
usage can vary greatly from one customer to the next.
Cloud billing automates the invoicing process for services provided online. This system is
particularly relevant for businesses that offer variable services, such as software-as-a-
service (SaaS), or that use any online subscription model. Here's how it works:

 Usage tracking: The system monitors the services that each customer uses. For instance,
in a SaaS model, it could track the number of users or the amount of data storage
consumed. This tracking ensures accurate, up-to-date billing data.
 Usage data processing: The cloud billing system processes this usage data according to
the business's pricing structure. Whether the model is pay-per-use, subscription or tiered,
the system applies the correct rates to the user data to calculate the bill.
 Automated invoicing: Based on these calculations, the system generates invoices
automatically. These invoices are then dispatched to customers, typically via email,
reducing manual intervention and the potential for billing errors.
 Integrated payment processing: Many cloud billing systems feature built-in payment
processing. This allows customers to pay their invoices online, offering convenience while
streamlining the payment collection process.
 Financial reporting: Finally, the system compiles detailed financial reports. These reports
give businesses insights into revenue patterns, outstanding payments and overall financial
health – helping you make informed decisions.
By handling the heavy lifting, cloud billing systems ensure that your billing is easy and
accurate, which is important for maintaining the fast pace(speed) of service that today's
customers expect.
Benefits of using cloud billing
Using cloud billing systems can bring several advantages to your businesses, including:
 Scalability: Cloud billing systems are designed to scale with a business's growth,
accommodating increases in transaction volume and complexity, without the need for
additional physical infrastructure.
 Cost savings: These systems typically operate on a subscription basis, which means that
businesses can avoid large up-front investments in hardware and software. This also helps
reduce expenditures associated with IT maintenance and upgrades.
 Flexibility: Cloud billing solutions offer customisable billing cycles, pricing models and
payment options to meet the demands of various customer segments.
 Accessibility: Being cloud-based, these systems provide the advantage of remote access –
allowing teams to handle billing operations from any location.
 Automation: Cloud billing systems allow you to automate routine tasks, such as
invoicing, payment processing and revenue recognition, which can lead to reduced errors
and administrative overheads.
 Integration: These platforms can typically integrate with other business systems – such as
customer relationship management (CRM) and enterprise resource planning (ERP) – to
maintain a cohesive (united or working together) data flow and improve the accuracy of
financial reporting.
 Compliance and security: Your cloud billing provider should be set up with financial
regulations and standards baked into the technology. Many also employ advanced security
measures to further protect sensitive financial data.
 Real-time analytics: Businesses can benefit from real-time insights into billing data,
which can inform strategic decisions and help identify trends and areas for optimisation.
 Customer experience: Cloud billing providers should have a user-friendly interface and
self-service portals, which can enhance the customer experience by providing transparency
and control over subscription and billing issues.
For businesses looking to update their billing infrastructure, a cloud-based platform can be
an agile, scalable option that's more economical than legacy systems. It may also come
with the added advantage of being managed by specialists, ensuring seamless operation
and up-to-date features.

What is Cloud Billing? Cloud billing refers to the process of managing costs associated with
cloud services. It includes tracking usage, managing subscriptions, and generating invoices
for services provided by cloud providers (e.g., AWS, Azure, Google Cloud).
Key Components

1. Pricing Models:
o Pay-as-you-go: You pay only for the resources you use.
o Reserved Instances: You commit to using a resource for a period (e.g., 1 or 3 years)
for a lower rate.
o Spot Instances: You bid for unused capacity at lower rates.

2. Usage Tracking:
o Detailed metrics on resource consumption (compute, storage, bandwidth) are
provided.
o Alerts and budgets can help manage costs effectively.

3. Billing Periods:
o Most providers bill monthly, but this can vary based on services and usage.

4. Invoices and Reports:

o Monthly invoices summarize charges, and usage reports help analyze spending
trends.
o Some platforms offer dashboards for real-time visibility.

5. Cost Management Tools:

o Tools and services are available for budget tracking, forecasting, and optimization
(e.g., AWS Cost Explorer, Azure Cost Management).

Best Practices

 Set Budgets: Define spending limits and receive alerts when nearing them.
 Regular Monitoring: Review usage and costs frequently to identify inefficiencies.
 Optimize Resources: Use tools to identify underutilized resources or consider switching to
more cost-effective options.

Conclusion

Effective cloud billing management helps organizations control costs, optimize resource
usage, and plan budgets accurately. Understanding your cloud provider’s billing structure is
crucial for making informed decisions.

What is load balancing?

Load balancing is the method of distributing network traffic equally across a pool of
resources that support an application. Modern applications must process millions of users
simultaneously and return the correct text, videos, images, and other data to each user in a
fast and reliable manner. To handle such high volumes of traffic, most applications have
many resource servers with duplicate data between them. A load balancer is a device that sits
between the user and the server group and acts as an invisible facilitator, ensuring that all
resource servers are used equally.
 What are the benefits of load balancing?

Load balancing directs and controls internet traffic between the application servers and their
visitors or clients. As a result, it improves an application’s availability, scalability, security,
and performance.

Application availability

Server failure or maintenance can increase application downtime, making your application
unavailable to visitors. Load balancers increase the fault tolerance of your systems by
automatically detecting server problems and redirecting client traffic to available servers.
You can use load balancing to make these tasks easier:

 Run application server maintenance or upgrades without application downtime

 Provide automatic disaster recovery to backup sites
 Perform health checks and prevent issues that can cause downtime

Application scalability

You can use load balancers to direct network traffic intelligently among multiple servers.
Your applications can handle thousands of client requests because load balancing does the
following:

 Prevents traffic bottlenecks at any one server

 Predicts application traffic so that you can add or remove different servers, if needed
 Adds redundancy to your system so that you can scale with confidence

Application security

Load balancers come with built-in security features to add another layer of security to your
internet applications. They are a useful tool to deal with distributed denial of service attacks,
in which attackers flood an application server with millions of concurrent requests that cause
server failure. Load balancers can also do the following:

 Monitor traffic and block malicious content

 Automatically redirect attack traffic to multiple backend servers to minimize impact
 Route traffic through a group of network firewalls for additional security

Application performance

Load balancers improve application performance by increasing response time and reducing
network latency. They perform several critical tasks such as the following:

 Distribute the load evenly between servers to improve application performance

 Redirect client requests to a geographically closer server to reduce latency
 Ensure the reliability and performance of physical and virtual computing resources

Disadvantages:

1. Complexity: Implementing load balancing in cloud computing can be complex,

especially when dealing with large-scale systems. It requires careful planning and
configuration to ensure that it works effectively.
2. Cost: Implementing load balancing can add to the overall cost of cloud computing,
especially when using specialized hardware or software.
3. Single Point of Failure: While load balancing helps to reduce the risk of a single point
of failure, it can also become a single point of failure if not implemented correctly.
4. Security: Load balancing can introduce security risks if not implemented correctly,
such as allowing unauthorized access or exposing sensitive data.

Admission Control in Cloud Computing

Admission control in cloud computing is a crucial mechanism that manages the allocation of
resources to various applications and services while ensuring that performance levels meet
predefined service quality requirements. Here are the key aspects:

1. Definition

Admission control is a process that determines whether a new request for resources (like
virtual machines, bandwidth, etc.) can be granted without violating the service level
agreements (SLAs) for existing workloads.

2. Objectives

 Resource Utilization: Optimize the use of available resources.

 Quality of Service (QoS): Ensure that performance metrics, such as latency and
throughput, are maintained.
 Scalability: Handle varying loads efficiently.

Types of Admission Control

 Static Admission Control:

 Determines resource allocation based on predefined thresholds.

 Does not adapt to current system conditions or dynamic workloads.

 Dynamic Admission Control:

 Continuously monitors system performance and adjusts resource allocation in real

time.
 Adapts to varying workloads and resource availability.

 Priority-Based Admission Control:

 Assigns different priority levels to tasks or users.

  Higher priority requests are granted access before lower priority ones.

 Resource Reservation:

 Allocates resources ahead of time for specific applications or users.

 Ensures guaranteed performance but may lead to underutilization.

 Load-Based Admission Control:

 Monitors system load and makes admission decisions based on current usage.
 Tries to balance the load across resources to prevent overloading.

 Elastic Admission Control:

 Dynamically scales resources up or down based on demand.

 Suitable for applications with fluctuating workloads.

 Policy-Based Admission Control:

 Uses predefined policies to determine admission based on various criteria, such as

user roles, service level agreements (SLAs), or application types.

 Queue-Based Admission Control:

 Implements a queuing system for incoming requests.

 Requests are processed based on a specific algorithm (e.g., FIFO, LIFO).

 Feedback-Based Admission Control:

 Uses feedback from ongoing processes to inform future admission decisions.

 Can optimize resource allocation based on historical performance data.

In cloud computing, static and dynamic admission control are two approaches to managing
resource allocation for incoming requests. Here’s a comparison of the two:

Static Admission Control

1. Predefined Policies: Resource allocation decisions are based on fixed, predetermined

rules or thresholds established before the system operates.
2. Limited Adaptability: Once set, the rules do not change in response to varying
workloads or real-time system performance.
3. Simplicity: Easier to implement and manage since it relies on straightforward criteria
for accepting or rejecting requests.
4. Performance Risk: Can lead to inefficiencies; for instance, underutilization of
resources during low demand or overloading during peak times.
5. Use Cases: Suitable for environments with predictable workloads where demand
patterns are stable.

Dynamic Admission Control

 1. Real-Time Decision Making: Resource allocation decisions are made based on real-
time system metrics, such as current load and performance.
2. High Adaptability: The system can adjust admission criteria dynamically,
responding to changing workloads and conditions.
3. Complexity: More complex to implement as it requires continuous monitoring and
sophisticated algorithms for decision-making.
4. Optimized Resource Utilization: Can enhance performance and efficiency by
admitting requests based on current capacity, potentially improving QoS.
5. Use Cases: Ideal for environments with variable workloads, such as public clouds
where demand can fluctuate significantly.

4. Techniques

 Threshold-based Policies: Requests are accepted or rejected based on resource

utilization metrics.
 Resource Reservation: Certain resources are reserved for specific applications to
ensure their performance.
 Load Balancing: Distributing workloads across resources to prevent bottlenecks.

5. Challenges

 Unpredictable Workloads: The variability in demand can make it difficult to predict

resource needs.
 Overhead: Implementing admission control can introduce additional processing
overhead.
 Trade-offs: Balancing between resource utilization and performance can be complex.

6. Implementation Considerations

 Monitoring Tools: Implementing robust monitoring to track resource usage and

performance.
 Policy Definition: Clearly defining SLAs and performance metrics.
 Feedback Mechanisms: Utilizing feedback from current workloads to adjust policies
dynamically.

7. Applications

 Cloud Service Providers: Ensure that they can offer reliable services to customers.
 Multi-tenant Environments: Manage resource allocation among various users to
prevent one tenant from monopolizing resources.

By effectively managing resource allocation through admission control, cloud computing

environments can maintain high availability, reliability, and performance for their users.

Automated policy-based management

Automated policy-based management in cloud computing refers to the use of predefined

policies to automate the management of resources, services, and applications. This approach
enhances efficiency, reduces manual intervention, and ensures compliance with
organizational standards. Here’s a deeper look into its key aspects:
1. Definition

Automated policy-based management involves creating rules or policies that dictate how
resources should be allocated, monitored, and managed within a cloud environment. These
policies can trigger automated actions based on specific conditions or thresholds.

2. Key Components

 Policies: These are rules that define acceptable behaviors and configurations. They
can cover aspects like security, resource allocation, performance, and compliance.
 Automation Tools: Software solutions or frameworks that execute policies without
human intervention. Examples include orchestration tools (orchestration is
the automated configuration, coordination, deployment, development,
and management of computer systems and software) and configuration management
system
 Monitoring Systems: Tools (like Sematext, The Elastic Stack, Prometheus, Zabbix
etc) that continuously assess the environment to ensure compliance with policies and
to trigger actions when necessary.

3. Benefits

 Consistency: Ensures that all resources are managed according to the same rules,
reducing variability.
 Efficiency: Automates repetitive tasks, freeing up human resources for more strategic
activities.
 Scalability: Policies can be applied consistently across a large number of resources,
making it easier to scale operations.
 Compliance and Security: Helps enforce compliance with regulatory standards and
security best practices.

4. Policy Types

 Resource Allocation Policies: Determine how and when resources are provisioned
and de-provisioned.
 Security Policies: Define access controls, authentication methods, and data protection
measures.
 Performance Policies: Set thresholds for resource usage, triggering scaling actions
when performance metrics are not met.
 Cost Management Policies: Help optimize resource usage to minimize costs, such as
shutting down unused resources.

5. Implementation Considerations

 Policy Design: Clearly define and document policies to ensure they meet
organizational goals.
 Monitoring and Feedback: Continuously monitor the outcomes of policies to refine
and improve them over time.
 Integration with Existing Tools: Ensure that policy management integrates
seamlessly with existing cloud management and orchestration tools.
6. Challenges

 Complexity of Policies: As the number of policies increases, managing and

maintaining them can become complex.
 Dynamic Environments: Cloud environments are often dynamic, which may require
frequent updates to policies.
 Balancing Automation and Control: Too much automation can lead to unintended
consequences (unpleasant result) if policies are not correctly defined.

7. Use Cases

 Resource Scaling: Automatically adjust the number of virtual machines based on

load.
 Security Enforcement: Automatically isolate or shut down resources that exhibit
suspicious behavior.
 Cost Optimization: Automatically adjust resource allocations based on usage
patterns to reduce costs.

Automated policy-based management is vital for organizations leveraging cloud services,

enabling them to maintain control, optimize resource usage, and ensure compliance in
increasingly complex environments.

LOAD BALANCING IN CLOUD

Load balancing in the cloud is a technique used to distribute workloads across multiple
computing resources to optimize resource use, minimize response time, and ensure high
availability.

What is Load Balancing?

Load balancing helps manage traffic and workloads by directing incoming requests to
multiple servers or resources based on various algorithms. It ensures no single resource
becomes overwhelmed, improving performance and reliability.

Types of Load Balancers

1. Software Load Balancers:

o Typically run on virtual machines (VMs) and offer flexibility.
o Examples: NGINX, HAProxy.

2. Hardware Load Balancers:

o Physical devices that manage traffic at a larger scale.
o Less common in cloud environments due to the rise of software solutions.

3. Cloud Provider Load Balancers:

o Managed services provided by cloud providers (e.g., AWS Elastic Load Balancing,
Google Cloud Load Balancing, Azure Load Balancer).
o Automatically scale and adapt to traffic demands.

Load Balancing Algorithms

 1. Round Robin:
o Distributes requests evenly across all servers in a circular order.

2. Least Connections:
o Directs traffic to the server with the fewest active connections, ideal for dynamic
workloads.

3. IP Hash:
o Uses the client’s IP address to determine which server handles the request, ensuring
that a client consistently connects to the same server.

4. Weighted Load Balancing:

o Assigns weights to servers based on their capacity, directing more traffic to stronger
resources.

Benefits of Load Balancing

 High Availability: Automatically reroutes traffic to healthy servers if one fails.

 Scalability: Easily add or remove resources based on demand.
 Performance Optimization: Distributes traffic efficiently, reducing latency.
 Fault Tolerance: Increases resilience by balancing loads and ensuring redundancy.

Use Cases

 Web Applications: Ensures smooth access to web servers, especially during high traffic.
 Microservices Architecture: Balances requests among multiple services in a distributed
environment.
 API Management: Handles incoming API requests effectively, ensuring quick response times.

Conclusion

Load balancing is essential for maintaining performance and reliability in cloud

environments. By intelligently distributing workloads, organizations can enhance user
experience, ensure high availability, and optimize resource usage. Implementing a well-
designed load balancing strategy is crucial for modern cloud applications.