CLOUD COMPUTING

Department of
Computer Science & Engineering

Department of Computer Science & Engineering www.cambridge.edu.in www.cambridge.edu.i

Department of Computer Science & Engineering www.cambridge.edu.i
 Chapters
• Module 1
‒ Chapter 1- Distributed System Models and Enabling Technologies
• Module 2
‒ Chapter 3: Virtual Machines and Virtualization of Clusters and Data Centers
• Module 3
‒ Chapter 4: Cloud Platform Architecture over Virtualized Datacenters
• Module 4
‒ Chapter 4(Texbook1) & Chapter 11 (Textbook 2)Cloud Security and Cloud
Security and Trust Management
• Module 5
‒ Chapter 6: Cloud Programming and Software Environments

Department of Computer Science & Engineering www.cambridge.edu.i

 What is cloud computing?
Cloud computing is a technological advancement it is
based on the concept of dynamic provisioning, which is
applied not only to services but also to compute capability,
storage, networking, and information technology (IT)
infrastructure in general. Resources are made available
through the Internet and offered on a pay-per-use basis
from cloud computing vendors.

Department of Computer Science & Engineering www.cambridge.edu.i

 SCALABLE COMPUTING OVER THE INTERNET

• Evolution of Computing Technology

• The Age of Internet Computing
• Platform Evolution
• High-Performance Computing (HPC)
• High-Throughput Computing (HTC)
• Emerging Computing Paradigms
• Computing Paradigm Distinctions
• Distributed System Families
• Future Computing Needs and Design Objectives

Department of Computer Science & Engineering www.cambridge.edu.i

 Future Computing Needs and Design Objectives

• Efficiency: Maximizing parallelism, job throughput, and power

efficiency.

• Dependability: Ensuring reliability and Quality of Service (QoS).

• Adaptation: Scaling to billions of requests over vast data sets.

• Flexibility: Supporting both HPC (scientific/engineering) and HTC

(business) applications.

Department of Computer Science & Engineering www.cambridge.edu.i

 Evolutionary trend toward parallel, distributed, and cloud
computing with clusters, MPPs, P2P networks, grids,
clouds, web services, and the Internet of Things.

It highlights
different
architectures
and
approaches
used for
distributed
and parallel
computing.

Department of Computer Science & Engineering www.cambridge.edu.i

 Scalable Computing Trends and New Paradigms
• Computing Trends and Parallelism
• Technological progress drives computing applications, as seen in Moore’s
Law (processor speed doubling every 18 months) and Gilder’s Law
(network bandwidth doubling yearly).
• Degrees of Parallelism (DoP):
• Bit-level (BLP),Instruction-level (ILP), Data-level (DLP), Task-level
(TLP), Job-level (JLP).
• Application domains
• Utility Computing and Cloud Adoption
• Hype Cycle of Emerging Technologies

Department of Computer Science & Engineering www.cambridge.edu.i

 Innovative Applications

Department of Computer Science & Engineering www.cambridge.edu.i

 Utility Computing and Cloud Adoption

• Utility Computing and Cloud Computing: Utility computing is a business model

where customers pay for computing resources.
• Grid and cloud computing act as utility service providers.
• Cloud computing extends beyond utility computing by running distributed
applications across edge networks.

Department of Computer Science & Engineering www.cambridge.edu.i

 Hype Cycle of Emerging Technologies
New technologies go
through five stages:

• Innovation trigger

• Peak of inflated
expectations

• Trough of
Disillusionment

• Slope of
Enlightenment

• Plateau of
Productivity

Department of Computer Science & Engineering www.cambridge.edu.i

 The Internet of Things and Cyber-Physical
Systems
• The Internet of Things (IoT)
• IoT extends the Internet to everyday objects, interconnecting
devices, tools, and computers via sensors, RFID, and GPS.
• History: Introduced in 1999 at MIT, IoT enables
communication between objects and people.
• Development & Challenges:
• IoT is in its early stages, mainly advancing in Asia and Europe.
• Cloud computing is expected to enhance efficiency, intelligence,
and scalability in IoT interactions.
Department of Computer Science & Engineering www.cambridge.edu.i
 Cyber-Physical Systems (CPS)
• CPS integrates computation, communication, and control
(3C) into a closed intelligent feedback system between the
physical and digital worlds.
• CPS enhances automation, intelligence, and interactivity in
physical environments.
• Still in the Trough of Disillusionment for full automation
(e.g., fully autonomous cars, widespread smart cities).

Department of Computer Science & Engineering www.cambridge.edu.i

 Technologies for Network Based Systems
• Introduction to Distributed Computing Technologies :
• Explores hardware, software, and network technologies for distributed
computing, focusing on distributed operating systems to manage massive
parallelism.
• Multicore CPUs and Multithreading Technologies:
• The growth of component and network technologies over 30 years has been
key to HPC and HTC development in terms of processor speed and network
bandwidth.
• Advances in CPU Processors: Includes multicore CPU Architecture,
processor Speed Growth, clock Rate Limitations, ILP Techniques for
Performance, GPU Parallel Processing.

Department of Computer Science & Engineering www.cambridge.edu.i

 Improvement in processor and network technologies over 33 years.

Department of Computer Science & Engineering www.cambridge.edu.i

 Architecture of a typical multicore processor

L1 cache is private to
each core, on-chip L2
cache is shared and L3
cache or DRAM Is off
the chip.

Department of Computer Science & Engineering www.cambridge.edu.i

 Multicore CPUs and Multithreading Technologies
Cont..

• Multicore CPU
and Many-Core
GPU
Architectures
• Multithreading
Technology

Department of Computer Science & Engineering www.cambridge.edu.i

 Multicore CPUs and Multithreading
Technologies Cont..
• GPU Computing to Exascale and Beyond
• How GPUs Work
• GPU Programming Model
• Power Efficiency of the GPU
GPU Programming Model

Department of Computer Science & Engineering www.cambridge.edu.i

 Example 1.1 the NVIDIA Fermi GPU Chip
with 512 CUDA Cores
• This is a streaming
multiprocessor (SM) module.
• Compute Unified Device
Architecture (CUDA) for parallel
computing on GPUs.
• Modern CUDA-based GPUs
leverage massive parallelism
with highly efficient memory and
thread scheduling, making
them ideal for HPC (High-
Performance Computing), AI,
deep learning, and scientific
simulations.
Department of Computer Science & Engineering www.cambridge.edu.i
 Power Efficiency of the GPU:

Department of Computer Science & Engineering www.cambridge.edu.i

 Memory, Storage, and Wide-Area Networking
• Memory Technology
• Disks and Storage Technology
• System-Area Interconnects
• Wide-Area Networking
• Virtual Machines and Virtualization Middleware
• VM Architectures
• Virtual Machines
• VM Primitive Operations
• Virtual Infrastructures

Department of Computer Science & Engineering www.cambridge.edu.i

 Disks and Storage Technology:
Improvement in memory and disk technologies over 33 years. The Seagate Barracuda
XT disk has a capacity of 3 TB in 2011.

Department of Computer Science & Engineering www.cambridge.edu.i

 System-Area Interconnects

• LAN connects users and servers for communication.

• NAS allows client hosts to access shared file-based storage via LAN.
• SAN connects servers with dedicated storage arrays using high-speed, block-level access.

Department of Computer Science & Engineering www.cambridge.edu.i

 Virtual Machines and Virtualization Middleware

Department of Computer Science & Engineering www.cambridge.edu.i

 VM Primitive Operations

Department of Computer Science & Engineering www.cambridge.edu.i

 Virtual Infrastructures

Department of Computer Science & Engineering www.cambridge.edu.i

 Data Center Virtualization for Cloud
Computing
• Data Center Growth and Cost Breakdown
• Low-Cost Design Philosophy
• Convergence of Technologies

Department of Computer Science & Engineering www.cambridge.edu.i

 SYSTEM MODELS FOR DISTRIBUTED AND CLOUD COMPUTING

Department of Computer Science & Engineering www.cambridge.edu.i

 Clusters of Cooperative Computers
• Cluster Architecture
• Single-System Image
• Hardware, Software, and Middleware Support
• Major Cluster Design Issues

Department of Computer Science & Engineering www.cambridge.edu.i

 Cluster Architecture:
• The architecture of a typical server cluster built around a low-latency, high bandwidth
interconnection network.
• Greg Pfister has indicated that an ideal cluster should merge multiple system images into a
single-system image (SSI).

Department of Computer Science & Engineering www.cambridge.edu.i

 Hardware, Software, and Middleware Support:

• Clusters using massive parallelism are known as Massively Parallel

Processors (MPPs) and dominate the Top 500 HPC list.
• Built from computer nodes (PCs, servers, SMPs), communication software
(PVM, MPI), and high-bandwidth networks (Gigabit Ethernet, Myrinet,
InfiniBand).
• OS & Middleware: Mostly run on Linux with cluster middleware to
support Single-System Image (SSI) and High Availability (HA).
• Parallel Applications: Need special environments for distributed resource
sharing, like Distributed Shared Memory (DSM).
• Virtual Clusters: Virtualization allows dynamic creation of clusters based
on user demand. (This means resources can be allocated dynamically,
depending on user demand, making clusters more flexible and efficient.)
Department of Computer Science & Engineering www.cambridge.edu.i
 Major Cluster Design Issues

• No unified cluster-wide OS for resource sharing.

• Middleware is required for cooperative computing and high
performance.
• Benefits of clusters: Scalability, efficient message passing, fault
tolerance, and management.

Department of Computer Science & Engineering www.cambridge.edu.i

 Grid Computing Infrastructures
• Computational Grids
• Grid Families

Department of Computer Science & Engineering www.cambridge.edu.i

 Grid Families

Department of Computer Science & Engineering www.cambridge.edu.i

 Peer-to-Peer Network Families
• P2P Systems
• Overlay Networks
• P2P Application Families
• P2P Computing Challenges

Department of Computer Science & Engineering www.cambridge.edu.i

 Overlay Networks

Department of Computer Science & Engineering www.cambridge.edu.i

 P2P Computing Challenges

Department of Computer Science & Engineering www.cambridge.edu.i

 Cloud Computing over the Internet
• Internet Clouds
• The Cloud Landscape

Department of Computer Science & Engineering www.cambridge.edu.i

 Internet Clouds
This diagram illustrates the concept of cloud computing and how users interact with cloud
services.

Department of Computer Science & Engineering www.cambridge.edu.i

 The Cloud Landscape

Department of Computer Science & Engineering www.cambridge.edu.i

 SOFTWARE ENVIRONMENTS FOR
DISTRIBUTED SYSTEMS AND CLOUDS
• Service-Oriented Architecture (SOA)
• Layered Architecture for Web Services and Grids
• Web Services and Tools
• The Evolution of SOA
• Grids versus Clouds
• Trends toward Distributed Operating Systems
• Distributed Operating Systems
• Amoeba versus DCE
• MOSIX2 for Linux Clusters
• Transparency in Programming Environments
• Parallel and Distributed Programming Models
• Message-Passing Interface (MPI)
• MapReduce
• Hadoop Library
• Open Grid Services Architecture (OGSA)
• Globus Toolkits and Extensions
Department of Computer Science & Engineering www.cambridge.edu.i
 Service-Oriented Architecture (SOA)
• Layered Architecture for Web
Services and Grids
• Web Services and Tools
• The Evolution of SOA
• Grids versus Clouds

Department of Computer Science & Engineering www.cambridge.edu.i

 Layered Architecture for Web Services and Grids
Layered Architecture for Web Services and Grids:
• Entity Interfaces: Define interfaces using WSDL (Web Services), Java methods, and CORBA IDL.
• Communication Systems: Utilize protocols such as SOAP (Web Services), RMI (Java), and IIOP
(CORBA), while supporting features like Remote Procedure Calls (RPC), fault recovery, and
specialized routing.
• Middleware Infrastructure: Utilizes message-oriented middleware such as WebSphere MQ or JMS
for message routing and virtualization.
• Fault Tolerance & Security: Web Services Reliable Messaging (WSRM) ensures message reliability.
Security measures include IPsec, secure sockets, and concepts from the OSI model.
• Discovery & Management Services: This includes JNDI, UDDI, LDAP, and ebXML for service
discovery. Service management is facilitated through CORBA Life Cycle, Enterprise JavaBeans, and
Jini’s lifetime model.
• Performance & Software Design: Distributed models provide high performance and software
modularity. There is a shift from CORBA and Java-based systems to modern architectures using
SOAP, XML, and REST.

Department of Computer Science & Engineering www.cambridge.edu.i

 The Evolution of SOA

Department of Computer Science & Engineering www.cambridge.edu.i

 Grids vs. Clouds

• Grid Computing:
• Uses static resources allocated in advance.
• Focuses on distributed computing with a defined structure.
• Cloud Computing:
• Uses elastic resources that scale dynamically.
• Supports virtualization and autonomic computing.

Department of Computer Science & Engineering www.cambridge.edu.i

 Trends toward Distributed Operating Systems

• Distributed Operating
Systems
• Amoeba versus DCE
• MOSIX2 for Linux Clusters
• Transparency in
Programming Environments

Department of Computer Science & Engineering www.cambridge.edu.i

 Transparency in Programming Environments

Department of Computer Science & Engineering www.cambridge.edu.i

 Parallel and Distributed Programming
Models
Distributed computing has four programming models with expected scalable
performance and application flexibility.
1. Message-Passing Interface (MPI)
2. MapReduce:
3. Hadoop Library
4. Open Grid Services Architecture (OGSA)
• Globus Toolkits and Extensions

Department of Computer Science & Engineering www.cambridge.edu.i

 Parallel and Distributed Programming
Models and Tool Sets

Department of Computer Science & Engineering www.cambridge.edu.i

 Open Grid Services Architecture (OGSA)

Department of Computer Science & Engineering www.cambridge.edu.i

 Performance, Security and Energy Efficiency
• Performance Metrics and Scalability Analysis
• Distributed system performance depends on several factors, including
CPU speed (MIPS), network bandwidth (Mbps), system throughput
(Tflops, TPS), job response time, and network latency.
• High-performance interconnection networks require low latency and
high bandwidth. Other key metrics include OS boot time, compile time,
I/O data rate, system availability, dependability, and security resilience.

Department of Computer Science & Engineering www.cambridge.edu.i

 Dimensions of Scalability
• Size Scalability – Increasing the number of processors, memory, or
I/O channels to improve performance.
• Software Scalability – Upgrading OS, compilers, and application
software to work efficiently in large systems.
• Application Scalability – Adjusting problem size to match machine
scalability.
• Technology Scalability – Adapting to hardware and networking
advancements while ensuring compatibility with existing systems.

Department of Computer Science & Engineering www.cambridge.edu.i

 Scalability versus OS Image Count

Department of Computer Science & Engineering www.cambridge.edu.i

 Performance Models: Amdahl’s and Gustafson’s Laws:

Department of Computer Science & Engineering www.cambridge.edu.i

 Problem with Fixed Workload:

Department of Computer Science & Engineering www.cambridge.edu.i

 Gustafson’s Law:
Let W be the workload in a given program. When using an n-processor system, the user
scales the workload to W′= αW +(1−α)nW.

Department of Computer Science & Engineering www.cambridge.edu.i

 Fault Tolerance and System Availability

• System Availability

Department of Computer Science & Engineering www.cambridge.edu.i

 Network Threats and Data Integrity
• Threats to Systems and Networks
• Security Responsibilities
• Copyright Protection
• System Defence Technologies
• Data Protection Infrastructure

Department of Computer Science & Engineering www.cambridge.edu.i

Department of Computer Science & Engineering www.cambridge.edu.i
 Loss of Confidentiality Loss of Integrity (Integrity
(Information Leakage) Violation)

• Eavesdropping: Unauthorized • Intercept/Alter: Manipulating data in

interception of communication. transit.
• Traffic Analysis: Analyzing data flow • Repudiation: Denying an action
patterns to gather intelligence.
• Penetration: Gaining unauthorized
• EM/RF Interception: Capturing access.
electromagnetic or radio frequency • Masquerade: Impersonating another
signals. user/system.
• Indiscretions of Personnel: Human • Bypassing Controls: Circumventing
errors that expose data. security mechanisms.
• Media Scavenging: Recovering deleted
or discarded data. • No Authorization: Accessing data without
permissions.
• Physical Intrusion: Gaining physical
access to tamper with data.

Department of Computer Science & Engineering www.cambridge.edu.i

 Loss of Availability (Denial of Improper Authentication
Service - DoS) (Illegitimate Use)
• DoS Attack: Flooding a system • Resource Exhaustion: Depleting
with traffic to make it system resources to cause a
unresponsive. crash.
• Trojan Horse: Malicious • Integrity Violation: Corrupting
software disguised as legitimate critical data, making services
software. unusable.
• Trapdoor: Hidden system • Theft: Stealing resources,
vulnerabilities allowing leading to unavailability.
unauthorized access.
• Replay Attacks: Repeating valid
• Service Spoofing: data transmissions to cause
Impersonating a legitimate disruptions.
service to mislead users.

Department of Computer Science & Engineering www.cambridge.edu.i

 Energy Efficiency in Distributed
Computing
Running a server farm is costly, requiring significant investments in
hardware, software, operations, and energy.
• Energy Consumption of Unused Servers
• Energy Management in Distributed Systems (Four Layers):
• DVFS Method for Energy Efficiency

Department of Computer Science & Engineering www.cambridge.edu.i

Department of Computer Science & Engineering www.cambridge.edu.i
 DVFS Method
for Energy
Efficiency

Department of Computer Science & Engineering www.cambridge.edu.i