CS-802-B Cloud Computing - May 2024 Question Paper

Under each question, How to answer lists the points an answer should cover, outlined by CampusPrep from the syllabus. Check the details against your textbook.

1. Marks
   aDraw a layered diagram to relate the construction of IaaS, PaaS and SaaS clouds from bare machine hardware to the user applications.
   How to answer

   • Layer 1: Bare machine hardware infrastructure (processors, memory, storage)
   • Layer 2: IaaS (virtual machines, computing resources, networking)
   • Layer 3: PaaS (middleware, runtime environments, development tools)
   • Layer 4: SaaS (applications and services for end-users)
   • Show how each layer builds upon the previous layer
   • Indicate progression from hardware resources to user applications
2. Marks
   bBriefly list the representative cloud service offerings at each cloud layer from the major cloud providers.
   How to answer

   • IaaS offerings: EC2, S3, Virtual Machines, Compute Instances, Storage Services
   • PaaS offerings: App Engine, Cloud Functions, Elastic Beanstalk, Platform services
   • SaaS offerings: Office 365, Salesforce, Google Workspace, Gmail
   • Mention at least 2-3 providers (AWS, Azure, Google Cloud)
   • Organize by layer with service names and brief descriptions
3. Marks
   aWhat virtualization support is needed to achieve fast cloning of VMs? Explain how VM cloning can enable fast recovery.
   How to answer

   • Copy-on-Write (CoW) technique for efficient cloning
   • Snapshot and clone functionality at hypervisor level
   • Shared storage and incremental copying mechanisms
   • VM cloning enables creation of identical VMs quickly
   • Fast recovery by reverting to previous clones or snapshots
   • Reduced RTO (Recovery Time Objective) and data loss prevention
4. Marks
   bWhat are hardware virtualization techniques? What are the benefits of virtualization.
   How to answer

   • Full virtualization: Complete abstraction of hardware, no host OS modification needed
   • Paravirtualization: Requires OS modification, better performance
   • Hardware-assisted virtualization: CPU support (Intel VT-x, AMD-V) for efficient virtualization
   • Benefits: Resource utilization, cost reduction, flexibility, scalability
   • Improved security, isolation between VMs, dynamic resource allocation
   • Enables live migration and fault tolerance
5. Marks
   aWhat is Bigtable? Describe the architecture of google file system.
   How to answer

   • Bigtable: Distributed storage system for structured data, sparse multi-dimensional sorted map
   • Built on top of GFS, uses HBase-like column families
   • GFS Architecture: Master (single), Chunkservers, Clients
   • Chunks of 64MB stored with replication factor typically 3
   • Master manages namespace, maintains file system tree, manages chunk leases
   • Clients interact with master for metadata, with chunkservers for data operations
6. Marks
   bWhat is MapReduce? How does MapReduce model differ from other models.
   How to answer

   • MapReduce: Programming model for processing large datasets in parallel
   • Consists of Map phase (applies function to input, produces key-value pairs) and Reduce phase (aggregates values by key)
   • Differs from MPI: MapReduce automatically handles distributed execution, fault tolerance, and data partitioning
   • Unlike message passing: No explicit communication between processes
   • Simpler abstraction for programmers, automatic load balancing and fault recovery
   • Suited for batch processing rather than streaming or interactive queries
7. Marks
   aDescribe various vulnerability assessment tool for cloud.
   How to answer

   • Cloud-specific vulnerability scanners: Nessus, Qualys, OpenVAS
   • Infrastructure assessment: AWS Config, Azure Policy for compliance checking
   • Penetration testing tools: Metasploit, Burp Suite adapted for cloud environments
   • Container security: Aqua Security, Twistlock for container vulnerability scanning
   • IaC scanning: Checkov, TerraForm scan for infrastructure-as-code vulnerabilities
   • Continuous monitoring and automated remediation capabilities
8. Marks
   bWhat is virtual threats? Explain VM specific security techniques.
   How to answer

   • Virtual threats: Security risks specific to virtualized environments (VM escape, side-channel attacks)
   • Hypervisor attacks: Compromise of hypervisor can affect all VMs
   • VM escape: Breaking out of VM boundary to access host or other VMs
   • Privilege escalation within VM or from VM to host
   • VM-specific security techniques: Hypervisor hardening, memory protection, CPU isolation
   • Secure boot, attestation, intrusion detection at hypervisor level, encrypted VM state
9. Marks
   aDescribe the shortcomings of mobile device and distance clouds for mobile cloud computing.
   How to answer

   • Mobile device shortcomings: Limited battery life, computational power, storage capacity
   • Intermittent and unreliable connectivity, variable network bandwidth
   • Distance cloud issues: High latency due to geographical distance
   • Increased network traffic and energy consumption for communication
   • Reduced quality of service (QoS) for real-time applications
   • Security and privacy concerns with remote data storage and transmission
10. Marks
    bWhat is cloud middleware? What are the importance of sky computing?
    How to answer

    • Cloud middleware: Software layer enabling communication and coordination between cloud services and applications
    • Provides abstraction, resource management, service discovery, and integration capabilities
    • Sky computing: Integration of cloud resources across multiple cloud providers (multi-cloud)
    • Importance: Avoids vendor lock-in, enables load balancing across clouds, improves fault tolerance
    • Leverages best services from different providers, cost optimization
    • Enhanced performance, scalability, and reliability through distributed processing
11. Marks
    aWhy are virtual machine and virtual clusters suggested in cloud computing system?
    How to answer

    • Virtual machines: Enable resource isolation, multiple OS on single hardware, efficient hardware utilization
    • VMs provide flexibility in resource allocation, dynamic scaling, and easier maintenance
    • Virtual clusters: Pool of VMs that behave as a single computing cluster
    • Improves fault tolerance through VM migration and live recovery
    • Cost efficiency: Better hardware utilization, reduced power consumption
    • Enables quick deployment, snapshots, cloning, and disaster recovery capabilities
12. Marks
    bWhat are the inter cloud issues? Describe functions of cloud plateforms.
    How to answer

    • Inter-cloud issues: Heterogeneity of APIs and standards across cloud providers
    • Data portability and migration challenges between clouds
    • Security and compliance concerns in multi-cloud scenarios, vendor lock-in risks
    • Cloud platform functions: Resource provisioning and management, service hosting and delivery
    • Monitoring, load balancing, auto-scaling, billing and usage tracking
    • Security management, disaster recovery, and SLA enforcement
13. Marks
    aDescribe relational database. Compare GFS and HDFS.
    How to answer

    • Relational database: Structured data storage using tables with rows and columns, ACID properties
    • GFS: Large-scale distributed file system by Google, optimized for large files, replication factor 3
    • HDFS: Hadoop File System, similar architecture to GFS, used in Hadoop ecosystem
    • Comparison: Both use master-slave architecture, replication for fault tolerance, support large files
    • GFS: 64MB chunks, HDFS: 128MB/256MB blocks; Different API and integration
    • GFS designed for Google infrastructure, HDFS for open-source Hadoop clusters
14. Marks
    bWhat are the issues in cloud computing? How to improve the performance of cloud plateforms?
    How to answer

    • Cloud computing issues: Security vulnerabilities, data privacy, vendor lock-in, compliance challenges
    • Latency and network bandwidth limitations, cost management and resource optimization
    • Performance improvements: Implement CDN for content delivery, geographic distribution of data centers
    • Load balancing, caching strategies, auto-scaling based on demand
    • Optimize VM placement, improve network protocols, containerization for better utilization
    • Regular monitoring, performance tuning, improved middleware solutions
15. Marks
    Write short notes on following (any two):

    1. a)Virtualized data center
    2. b)Multitenant techniques
    3. c)Load balancing and Resource optimization
    How to answer

    • Choose and explain any two from the three options
    • Virtualized data center: Pooled resources, VM management, automated provisioning
    • Multitenant techniques: Logical isolation, resource sharing, customization per tenant
    • Load balancing: Distribution of workload across servers, performance optimization
    • Resource optimization: Efficient CPU/memory/storage utilization, cost reduction