Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Google Cloud Platform, Building cost-effective, flexible and scalable education resources

Google Cloud Platform, Building cost-effective, flexible and scalable education resources

Google Apps for Education has helped transform the way that technology is used to drive collaborative and flexible learning in Higher Education. Google Cloud Platform which includes App Engine, Compute Engine and BigQuery can be tightly integrated into any existing deployment of Google Apps and allows organisations to build custom functionality and services using the same underlying foundations that power Google’s own services. This talk will include a number of case studies to explore how you can build next generation teaching services on Google Cloud Platform to compliment existing use of Google Apps and Chromebooks. The talk will also explore the use of App Engine and Compute Engine to deliver highly cost-effective, flexible and scalable virtual computing labs.


July 16, 2014

More Decks by sharifsalah

Other Decks in Technology


  1. Sharif Salah, University of Portsmouth Google Developer Expert 2014 Google+

    profile @sharif_salah about.me/sharif.salah Google Cloud Platform Building cost-effective, flexible and scalable education resources
  2. Image courtesy of Google/Connie Zhou

  3. Agenda Cloud Platform overview Existing usage in education Potential future

    usage Resources Q&A
  4. Cloud Platform overview

  5. Why Google Cloud Platform? Tight integration with Google Apps Rich

    variety of cutting edge PaaS & IaaS services Build on the same infrastructure that powers Google! Hands-off operation and monitoring including auto-scaling
  6. Compute Compute Engine App Engine

  7. Storage Cloud Datastore Cloud Storage Cloud SQL

  8. Services Cloud Endpoints BigQuery

  9. Cloud Console Cloud SDK REST APIs

  10. IaaS PaaS SaaS Gmail, Google Drive and Google Calendar App

    Engine Compute Engine Cloud Storage
  11. The story so far

  12. App Engine PaaS Java, Python, PHP & Go Free quota

    Auto scaling Massively scalable Traffic splitting
  13. Cloud Storage IaaS Multi-petabyte Globally replicated Tightly integrated into GCP

    Flexible security model
  14. myport.ac.uk (students & alumni) port.ac.uk (staff) Google Apps domains

  15. App Engine & Datastore Middleware Web traffic Student records Assessments

    (read only)
  16. Cloud Storage Middleware Web traffic Document Store Documents / Static

    Web Content
  17. Lessons learnt Billing / support is really important to address

    early on User authentication is super-simple for Google Apps domains Web apps require a small amount of ongoing development The platform is growing very rapidly In practise App Engine is super-reliable
  18. Support Choice of Silver, Gold or Platinum Can then apply

    for Premier status including offline billing Architecture Support Google Enterprise Support Center Optional 24x7 and Phone support
  19. myport.ac.uk (students & alumni) port.ac.uk (staff) Google Apps domains gcp.port.ac.uk

    (GCP admins) Used to manage all Google Cloud Platform projects with billing enabled under a single contract
  20. Where to next?

  21. Compute Engine IaaS Compute, Storage & Network Sub-hour billing Hugely

    scalable Consistent performance Linux (& Windows preview)
  22. Usage outline Micro (shared CPU) instances ideal for mass education

    use One virtual machine per student One hour lab usage for 30 students and instances less than $0.50 One year hosting approx $90.00 per (micro) instance Use App Engine and APIs to orchestrate
  23. App Engine Compute Engine Orchestration (or use Ansible, Chef, Puppet

    or Salt) Virtual machine running Node.js, PHP, Python ...etc. Student / User 1 1 1…*
  24. Lightweight containers Multiple containers per instance Transfer workloads between local

    infrastructure and cloud!
  25. “Kubernetes builds on top of Docker to construct a clustered

    container scheduling service. Kubernetes enables users to ask a cluster to run a set of containers. The system will automatically pick worker nodes to run those containers on, which we think of more as "scheduling" than "orchestration". Kubernetes also provides ways for containers to find and communicate with each other and ways to manage both tightly coupled and loosely coupled sets of cooperating containers.” Source: https://github.com/GoogleCloudPlatform/kubernetes/blob/master/DESIGN.md Kubernetes
  26. App Engine Orchestration (or use Ansible, Chef, Puppet or Salt)

    Compute Engine Instance + Docker containers Student / User C1 C2 Cn ..
  27. Caveats Port 25 blocked for outgoing mail (use Sendgrid or

    similar) TCP, UDP or ICMP traffic only By default only port 22 (SSH) is open By default SSH password authentication is disabled Check for default quota limits in all services
  28. Summary We use Cloud Storage to expose highly-reliable, user-facing static

    web content Use App Engine to host auto-scaling, dynamic & high-availability web services Future use of Compute Engine for education resources and web services Access to cutting edge technology that frees us up to focus on core business
  29. Resources

  30. None
  31. Case studies Compliance information Pricing calculator Research Support packages Further

    reading See also Compute Engine Management with Puppet, Chef, Salt, and Ansible Containers on Google Cloud Platform Docker Integrating Google Compute Engine and Ansible
  32. Google Cloud Platform would like to offer all participants of

    IWMW 2014 $500 of credit to build your web or mobile app! To get started, follow the three steps below: 1. http://g.co/cloudstarterpack 2. Click "apply now" 3. Use the code: gde-in
  33. These slides: http://goo.gl/No9gFM Thank you! Any questions?