Unit 4

Cont…
Types of clouds
Four different types of cloud:
• Public clouds.
• Private clouds.
• Hybrid or heterogeneous clouds.
• Community clouds.
Public clouds
Public clouds are the first expression of cloud computing, offering services to anyone, anywhere, and anytime
via the Internet.

• They are a distributed system, likely composed of one or more datacenters connected together, on top of
which specific cloud services are implemented.

• Historically, public clouds were the first class of cloud, offering solutions for minimizing IT infrastructure costs
and handling peak loads on local infrastructure.

• They are attractive for small enterprises, allowing them to start their businesses without large up-front
investments.

• Public clouds allow dynamic upsizing or downsizing of IT according to business needs.

• They are used to replace or extend enterprise IT infrastructure when needed.

• A fundamental characteristic of public clouds is multitenancy, serving a multitude of users, not a single
customer.

• QoS management is a crucial aspect of public clouds, with significant software infrastructure devoted to
monitoring cloud resources, billing them according to contracts, and keeping a complete history of cloud usage
for each customer.
• Offers various services: infrastructure, platform, or applications.

• Examples include Amazon EC2, Google AppEngine, and SalesForce.com.

• Characterized by their ability to scale on demand and sustain peak loads.

• No restrictions on the type of distributed system used to support public clouds.

• Typically, one or more datacenters form the physical infrastructure.

• Public clouds can be geographically dispersed to share user load and better serve them.

• Amazon Web Services offers three different regions: us-west-1, us-east-1, and eu-west-1.

• Different software stacks are installed for different classes of services.

Private clouds
• Public Clouds and Private Clouds: A Comparative Analysis
• Public Clouds: A Cost-Effective Solution
• Public clouds offer cost reduction and reduced capital expenses.
• Critics argue that the provider controls the infrastructure and customer's sensitive data.
• Institutions like government and military agencies may not consider public clouds for
sensitive data processing or storage.
Loss of Control: Potential Risks
• Geographical location of a datacenter can affect regulations on digital information
management.
• Sensitive information can be made accessible to government agencies or considered
illegal if processed with specific cryptographic techniques.
• Existing enterprises with large computing infrastructures may not want to switch to
public clouds.
Private Clouds: A Solution
• Private clouds, similar to public clouds, provide dynamic provisioning of computing
resources.
• They rely on a private infrastructure and keep core business operations in-house.
• Security concerns are less critical as sensitive information doesn't flow out of the private
infrastructure.
key advantages of using a private cloud computing
infrastructure:
• Provides customer information protection: In-house security is easier
to maintain and rely on than public cloud offerings.
• Ensures infrastructure ensuring Service Level Agreements (SLAs):
Quality of service requires specific operations like clustering, failover,
data replication, system monitoring, maintenance, and disaster
recovery.
• Ensures compliance with standard procedures and operations:
Organizations subject to third-party compliance standards need
specific procedures for application deployment and execution, which
virtual public infrastructure may not provide.
Hybrid clouds
Hybrid Clouds: A Combination of Private and Public Clouds
• Public clouds are large software and hardware infrastructures, but they
face security threats and administrative issues.
• Private clouds are ideal for maintaining information processing within an
enterprise or using existing hardware and software infrastructure.
• Private deployments struggle with scaling on demand and efficiently
addressing peak loads.
• Hybrid clouds leverage the capabilities of both private and public clouds,
allowing enterprises to exploit existing IT infrastructures and maintain
sensitive information.
• Hybrid clouds naturally grow and shrink by provisioning external resources
and releasing them when needed.
• Security concerns are limited to the public portion of the cloud, which can
perform operations with less stringent constraints.
• Hybrid clouds are heterogeneous distributed systems resulting from a
private cloud that integrates additional services or resources from one or
more public clouds.
• Hybrid Cloud Concept and Management
• Hybrid cloud primarily applies to IT infrastructure, not software services.
• Service-oriented computing integrates paid software services with existing
applications.
• Dynamic provisioning in IaaS involves acquiring on-demand virtual machines to
increase the capability of the distributed system.
• Infrastructure management software and PaaS solutions are key for deploying
and managing hybrid clouds.
• Dynamic provisioning introduces a complex scheduling algorithm and policies
for private clouds, aiming to optimize the budget spent on public resources.
• OpenNebula and InterGrid are two approaches to managing virtual machine
allocation in public clouds.
• OpenNebula integrates advanced schedulers like Haizea for cost-based
scheduling.
• InterGrid is a distributed scheduling engine that manages virtual machine
allocation in peer networks, considering user budget and peering arrangements.
Community clouds

• Community clouds are distributed systems created by integrating the

services of different clouds to address the specific needs of an industry, a
community, or a business sector.
• The infrastructure is shared by several organizations and supports a
specific community that has shared concerns (e.g., mission, security
requirements, policy, and compliance considerations). It may be managed
by the organizations or a third party and may exist on premise or off
premise.
• Community Cloud Usage Scenario
• Users of community clouds are identified communities with shared
needs.
• Users can be government bodies, industries, or simple users.
• Community clouds differ from public clouds, which serve multiple
users with different needs.
• Community clouds differ from private clouds, where services are
typically delivered within the cloud owner's institution.
• Community Clouds in Sectors:

• Media Industry: Community clouds can improve content production efficiency by providing a
shared environment for business-to-business collaboration. They offer the necessary bandwidth,
CPU, and storage for efficient media production.

• Healthcare Industry: Community clouds can provide a global platform for sharing information
and knowledge without revealing sensitive data. They can support the storage of patient-related
data in a private cloud while automating processes within hospitals.

• Energy and Core Industries: Community clouds can bundle comprehensive solutions for
management, deployment, and orchestration of services and operations. They can create an open
and fair market.

• Public Sector: Community clouds can provide a distributed environment for strategic solutions
at local, national, and international administrative levels. They can facilitate
business-to-administration, citizen-to-administration, and business-to-business operations.

• Scientific Research: Community clouds are ideal for organizations sharing a large distributed
infrastructure due to scientific computing.
• Community Cloud Benefits

• Openness: Removing dependency on cloud vendors allows fair

competition.
• Community: Infrastructure is scalable due to collective provision of
resources and services.
• Graceful Failures: No single provider or vendor controls the
infrastructure, ensuring no single point of failure.
• Convenience and Control: Shared and owned by the community,
decision-making is collectively democratic.
• Environmental Sustainability: Harnesses underutilized resources,
reducing carbon footprint.
• Organic Growth: Supports community demand, sustaining it.
Economics of the cloud

• The main drivers of cloud computing are economy of scale and

simplicity of software delivery and its operation. In fact, the biggest
benefit of this phenomenon is financial: the pay-as-you-go model
offered by cloud providers. In particular, cloud computing allows:
• Reducing the capital costs associated to the IT infrastructure
• Eliminating the depreciation or lifetime costs associated with IT
capital assets
• Replacing software licensing with subscriptions
• Cutting the maintenance and administrative costs of IT resources
• Capital Costs and IT Infrastructure
• Capital costs are one-time expenses that contribute to long-term profit.
• IT infrastructure and software are capital assets as they are required for
business operations.
• IT resources, including payroll, customer relationship management,
enterprise resource planning, and product tracking, are capital assets for
any enterprise.
• Depreciation costs, which represent the loss of hardware value over time
and the aging of software products, are significant expenses.
• Cloud computing reduces these costs by shifting capital costs into
operational costs inducted by renting infrastructure and paying software
subscriptions.
• Cloud computing also reduces administrative and maintenance costs,
eliminating the need for administrative staff and IT support staff.
• Depreciation costs disappear as all IT needs are served by the cloud,
eliminating the need for IT capital assets that depreciate over time.
• Cloud Computing Pricing Models

• Tiered Pricing: Cloud services offered in multiple tiers with fixed

computing specifications and SLAs at a specific price per unit of time.
• Per-Unit Pricing: Revenue source determined in units of specific
services like data transfer and memory allocation.
• Subscription-Based Pricing: Mostly used by SaaS providers where
users pay a periodic subscription fee for software or specific
component services integrated in their applications.
Open challenges
Cloud interoperability and standards

Cloud Computing: Service-Based Model and Vendor Lock-In

• Cloud computing is a service-based model for IT infrastructure and
applications.
• Standards and interoperability between vendors are crucial for
successful cloud computing adoption.
• Vendor lock-in is a significant barrier, especially for enterprises with
significant IT revenue.
• This can prevent customers from switching to competitors' solutions,
leading to significant costs and time.
• Implementing and adopting standards in the cloud computing
community can reduce vendor lock-in risks.
• Vendor lock-in due to proprietary virtual machine format.
• Open Virtualization Format (OVF) proposed for storing information
and metadata about virtual machine images.
Scalability and fault tolerance

• Cloud middleware must be designed for scalability in performance,

size, and load.
• Cloud middleware manages a large number of resources and users,
requiring significant administrative and maintenance costs.
• Tolerating failure is crucial in cloud systems, often more important
than efficient and optimized systems.
• The challenge lies in designing scalable, fault-tolerant systems that
are easy to manage and provide competitive performance.
Security, trust, and privacy

• Massive use of virtualization technologies exposes existing systems to

new threats.
• Cloud applications can process sensitive information, stored in
advanced cryptography technology.
• Data processed in memory must be decrypted by the legitimate
application, but can be accessed by a virtual machine manager.
• Lack of control over the application execution environment creates
new security threats.