1. Design and Implementation of the Workflow of an Academic Cloud Abhishek Gupta, Jatin Kumar, Daniel J Mathew, Sorav Bansal, Subhashis Banerjee, Huzur Saran IIT Delhi

2. Introduction  Why move to the cloud?  Lower cost  Higher utilization of computing resources  Ease of management  Why have a private cloud?  Privacy  Security Design and Implementation of the Workflow of an Academic Cloud 2

3. A Typical Academic Scenario  IIT Delhi  8000 students  450 faculty members  1000+ workstations  Many labs and server rooms across campus  Issues  High costs: hardware, power, management  Required: consolidation Design and Implementation of the Workflow of an Academic Cloud 3

4. Requirements for an IaaS Academic Cloud  Workflow: should be simple and usable by all  Cost and configurability: ideal if open source  Performance: both compute and I/O performance are required  Maximize resource utilization Design and Implementation of the Workflow of an Academic Cloud 4

5. Currently Available Private Cloud Solutions  Ubuntu Enterprise Cloud  Familiarity with both Ubuntu and Eucalyptus required  Documentation is insufficient  Not highly customizable  Eucalyptus does provide a API …  VMWare vCloud  One of the most stable cloud platforms, but…  Licensing costs are very high  These private cloud solutions are for enterprise use; what about an academic setup? Design and Implementation of the Workflow of an Academic Cloud 5

6. Baadal: Overview  Baadal = cloud (in Hindi)  Technology stack Design and Implementation of the Workflow of an Academic Cloud 6 BaadalWeb2pyLibvirtHypervisor LayerBlade Server

7. Baadal: Components Design and Implementation of the Workflow of an Academic Cloud 7

8. Baadal: Workflow Step 1: User logs in Design and Implementation of the Workflow of an Academic Cloud 8

9. Baadal: Workflow Step 2: User gives the specifications of the VM: name, operating system template and RAM. He also specifies a faculty member who will approve this request. Design and Implementation of the Workflow of an Academic Cloud 9

10. Baadal: Workflow Design and Implementation of the Workflow of an Academic Cloud 10 Step 3: Faculty member logs in. He approves/rejects the requests that have queued up till then. This request now goes to the data center administrators who will give the final approval.

11. Baadal: Workflow Step 4: Data center administrator logs in. He appropriately approves, modifies or rejects the requests approved by faculty. Design and Implementation of the Workflow of an Academic Cloud 11

12. Baadal: Workflow Step 5: Once the request is approved, an appropriate host is selected. Step 6: A pre-installed template with the required OS and software is deployed to the selected host. A new port on the perimeter server is forwarded to the VNC port of the newly created VM. A VNC password is also configured. Design and Implementation of the Workflow of an Academic Cloud 12

13. Baadal: Workflow Design and Implementation of the Workflow of an Academic Cloud 13 Step 7: The connection details (IP address, VNC port and password) are sent to the user.

14. Cost Model and over-provisioning  Gold (1:1), Silver (1:2), Bronze (1:4) and Shutdown  Users are expected to switch levels while the machine is running; no reboot required  Users are charged accordingly Design and Implementation of the Workflow of an Academic Cloud 14

15. Scheduler  On runlevel change, a VM is live-migrated to an appropriate host  Greedy scheduling strategy used to find new host  On spawning/starting a VM our algorithm assigns an appropriate host Design and Implementation of the Workflow of an Academic Cloud 15

16. Perimeter server  Fixed IP:port combination not possible due to migrations  Perimeter server maps user ports to actual IP:port combinations using iptables to do packet forwarding Design and Implementation of the Workflow of an Academic Cloud 16

17.  With granularity of day, week and month to influence user’s choice of runlevel Performance Graphs Design and Implementation of the Workflow of an Academic Cloud 17

18. Other Features  Wake-On-Lan  Task Queues  Support for multiple datastores  Automated emails/reminders  Storage-specific plugins to optimize operations Design and Implementation of the Workflow of an Academic Cloud 18

19. Baadal: Infrastructure  32 blade servers each with 2x6 core Intel® Xeon® CPU X5670 @ 2.93GHz and 16 GB RAM  16 blade servers each with 2x4 core Intel® Xeon® CPU E5540 @ 2.53GHz and 12 GB RAM  A 10Gbps Ethernet backbone  50 TB of virtualized storage based on a NetApp 3210V NAS and HP EVA6400 SAN with FC disks Design and Implementation of the Workflow of an Academic Cloud 19

20. Baadal: Performance Test 1 KVM + Desktop 2 KVM + Server 3 VMWare + Server 4 Empty loop (10,000,000) 21840 us44321 us44553 us Fork (1,000,000)29.72s6.88s3.97s wget (685.29MB)54.09s20.36s9.5s cp (685.29MB)71.97s11.65s26.07s iscp (685.29MB)29.64s52.34s4.75s oscp (685.29MB)73.54s83.68s4.86s Ping hypervisor0.2886s0.3712s0.1204s Design and Implementation of the Workflow of an Academic Cloud 20 1.Each VMs is allocated 1GB RAM, 1 vCPU and a 10 GB hard disk. 2.Desktops used are lab machines with typical configuration as 4GB RAM, C2D, 500GB hard disk on a 1Gbps Ethernet 3.KVM + Server refers to KVM hypervisor running on HP Proliant BL460cG7 (16GB RAM, 24 CPU, 10Gbps Ethernet) 4.VMWare + Server refers to VMWare as hypervisor running on Dell PowerEdge R710 (24GB RAM, 16 CPU, 10Gbps Ethernet)

21. Changes to Baadal [by CSC@IITD]  We are not handing over the port number for connecting to the VNC server due to network traffic issues and also most of the Linux users can use SSH and Windows users can use Remote Desktop from inside Windows. But we have kept VNC ports for administration purposes.  We are using Kerberos for back-end authentication; it can be easily replaced by any other authentication mechanism using inbuilt Web2py authentication modules.  Faculty/students in the admin team have two roles i.e. normal user and administrator, which they can switch between. Design and Implementation of the Workflow of an Academic Cloud 21

22. Future Work: Community-based Cloud Design and Implementation of the Workflow of an Academic Cloud 22  Can be used to put under-utilized resources like lab machines to good use  Challenges  Reliability  Network and storage topology  Scheduling  Solution for high reliability: record-replay  Already implemented successfully in Linux/KVM

23. References 1. Laor Kivity, Kamay, Lublin, and Liguori. KVM: The Linux Virtual Machine Monitor. Virtualization Technology for Directed I/O. Intel Technology Journal, 10:225-230, July 2007. 2. Libvirt, the virtualization API. http://www.libvirt.org.http://www.libvirt.org 3. Massimo DiPierro. Web2py Enterprise Web Framework, 2 nd Ed. Wiley Publishing, 2 nd edition, 2009. 4. Ubuntu enterprise cloud - overview. http://www.ubuntu.com/business/cloud/overview. http://www.ubuntu.com/business/cloud/overview 5. VMWare vCloud Director - deliver infrastructure as a service without compromise. http://www.vmware.com/products/vcloud- director/features.html. http://www.vmware.com/products/vcloud- director/features.html Design and Implementation of the Workflow of an Academic Cloud 23

24. Thank You Questions?