Your SlideShare is downloading. ×
0
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Federated Cloud Computing
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×
Saving this for later? Get the SlideShare app to save on your phone or tablet. Read anywhere, anytime – even offline.
Text the download link to your phone
Standard text messaging rates apply

Federated Cloud Computing

5,421

Published on

Cloud Computing has seen phenomenal growth in the last few years. Although large enterprises see a lot of gaps like vendor lock-in with both Public Cloud offerings as well as Private Clouds leading to …

Cloud Computing has seen phenomenal growth in the last few years. Although large enterprises see a lot of gaps like vendor lock-in with both Public Cloud offerings as well as Private Clouds leading to slower adoption than expected. Federated Clouds may be able to solve the problem. Cloud Federation brings together cloud resources from public as well as private environments so that enterprises can choose an environment on-demand that makes sense for their particular workloads across vendors. While doing so, it also provide a unified access and view of these resources allowing enterprises to easily manage, monitor and meter them. The aim of the presentation is to understand the need of a federated cloud and a reference architecture to go about building one.

Published in: Technology, Business
0 Comments
8 Likes
Statistics
Notes

  • Be the first to comment

Show More
No Downloads
Views
Total Views
5,421
On Slideshare
0
From Embeds
0
Number of Embeds
4
Actions
Shares
0
Downloads
0
Comments
0
Likes
8
Embeds 0
No embeds
Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide

Transcript

  • 1. Cloud Federated Clouds: Computing The Future of Cloud Computing for DevOps Storage Enterprises Enterprise &Scalable Apps Data Analytics Chirag Jog CTO, Clogeny Technologies [email protected]
  • 2. Agenda Cloud services in a snapshot Economic Theory Understanding Cloud Federation Example of a federated Multi-tier Application Reference Architecture for Federated Clouds (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 3. Cloud Services in a Snapshot SaaS PaaS IaaS (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 4. Economic Theory Vendor Lock-in Hold-up Problem and Under-investment (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 5. Vendor Lock-in Vendor Lock-in • Uncertainty of selecting an unknown technology • Learning curve of a technology Higher vendor lock-ins in PaaS • Google App Engine, Force.com Lower vendor lock-ins in IaaS • AWS, Rackspace This leads to the consequence, that users depend on the business strategy of the service provider. (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 6. Hold-up Problem and Under-investment Hold Up Problem: • Bargaining Power has shifted to the provider: He can negotiate higher fees. Under-investment: • Risk of Lock-in • Anticipation that Bargaining Power will shift. A party being forced to accept sub-optimal conditions cannot escape the situation due to the lock-in and �?nds itself in a hold-up. Example: Closed PaaS (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 7. Motivation for Cloud Federation No* Vendor lock-in Multitude of Players Heterogeneity of cloud platforms Interoperability and Portability Geographical distribution (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 8. Designing a Federated Cloud Services from different providers aggregated in a single pool supporting three basic interoperability features: • Resource migration • Resource redundancy • combination of complementary resources & services (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 9. Redundancy versus Migration Redundancy: is used whenever there is a subset of (properly organized) service offerings that provide better utility to a client than any single service offering Migration: can be triggered when a new service offering offers better utility to a client than any previously used service offering (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 10. Resource Redundancy IaaS:  Compute Resources – Redundant Deployment – Redundant Computation – Parallel Computation  Storage Resources – Replication – Fragmentation (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 11. Resource Redundancy Redundant Deployment: • Application logic is deployed to different providers. • Incoming requests are processed by only one instance. • Increase the availability while decreasing provider dependence. • Compliance with regulations, which require instances. • Customer proximity could be an issue to reduce latency • Automatic Failover (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 12. Resource Redundancy: Redundant Deployment (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 13. Resource Redundancy Redundant Computation: • Application logic is deployed to different providers • Every request is processed by more than one instance • Improve performance by reducing the risk of an instance failing right before completing a task Parallel Computation: • the data is broken down at bit level and processed at different providers’ sites following the same application logic or complimentary services are deployed to different providers. (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 14. Resource Redundancy: Redundant Computation (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 15. Resource Redundancy: PaaS PaaS: • Hard to use redundantly :  As they usually follow a different programming model  Support only a limited number of programming languages  Limitations on the programming model they build upon so that applications need to be fine-tuned for a particular platform.  Example: Google App Engine, Force.com • Open source PaaS:  Hosted by the customer on top of IaaS  Example: VMWare Cloud Foundry, Redhat Open Shift (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 16. Resource Redundancy: SaaS SaaS: • Very hard and probably cost-intensive to federate complex services like SalesForce. • Virtually impossible to isolate smaller building blocks of the service as no competing solutions exist, which offer exactly the same functionality. • The potentially proprietary data formats and APIs of such services increase the problem. (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 17. Resource Migration - 1 Migration incorporates scenarios where resources are being transferred from one Cloud provider A to another Cloud provider B. • Shadowed or redundant migration • Non-redundant migration • Full Migration • Partial migration (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 18. Resource Migration - 2 (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 19. Towards a Reference Architecture - 1 Interoperable systems – • Compatible Cloud Management Systems  Eucalyptus, Amazon EC2, Cloudstack Cloudbridge • Multi-Cloud Libraries  Libcloud – Python,  jCloud – Java  Fog – Ruby (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 20. Towards a Reference Architecture – 2 Provisioning Engine • Discover application packages and components. • Map Business components to pool of resources. • Generate Deployment descriptors based on user requirements and existing discovered resources and policies (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 21. Towards a Reference Architecture - 3 Resource Manager • Manage all the resources in a unified way • Provide an abstraction of the APIs of the underlying services • Allow configurations of resources in unified way using adaptors like multi-cloud libraries (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 22. Towards a Reference Architecture - 4 Deployment Manager • Based on a deployment description the manager executes resource management commands through Resource Managers. • It guarantees the availability and the correct configuration of provisioned resources. • Performs transformation of images across virtualized environments. • Provide Guarantees and SLAs (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 23. Towards a Reference Architecture - 5 Configuration Manager: • Recreate Virtual Appliances and application packages based on existing configurations. Data Distribution Manager: • Setup data storage as per configuration - Data replication, Data redundancy, Data Sharding Monitoring Systems: • Gather information about resource states • Report unexpected conditions (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 24. Reference Architecture Business Logic Policy Provisioning Engine Multi – Cloud Resource cloud Cloud Manager Service Cloud Deployment libraries Service Provider Service Provider Manager Transformation Provider Configuration Engine Manager Monitoring System Multi – Resource Data cloud Manager libraries Distribution Manager (US) 408-556-9645Clogeny Technologies (India) +91 20 661 43 482 http://www.clogeny.com
  • 25. Thanks ! Cloud Cutting Edge Computing Fun @ DevOps Work Storage Enterprise &Scalable Apps Data [email protected] Analytics

×