Dear Alan and occi-wg, SAJACC is a part of NIST cloud computing program. One of the activities of SAJACC is to document cloud computing *technical* use cases. The first version of the draft use case document was released last November [1]. These use cases are meant to document small units of work of how a cloud subscriber interacts with the cloud. By looking at cloud interfaces out there and see if these interfaces support realizing these requirements, we hope to surface gaps and gain insights into cloud interface standards. The testing we have been doing is mostly "feasibility" testing: answering questions like "can this be done using interface X". The initial set of test code and observations are available on SAJACC collaboration site [2]. With regard to VM control related use cases, we have been writing Java client applications ("testing code"), and running them against the interface endpoint to produce a output to show (hopefully) the requirements documented in the VM control use cases can (or can not) be met by interface X. The VM allocation use case (3.7) for example, says: (1) The cloud-subscriber requests a specific pre-defined Virtual Machine image supplied by the cloud-provider (O/S, CPU cores, memory, and security) and launches new VM instances and (3) The cloud-subscriber has secure launching and administration of their VM instance. We believe these are reasonable requirements by a lot of cloud (IaaS) consumers. The VM state management use case (3.8) for example, documents that a subscriber would like to have the ability to persist VM states, perhaps to a persistent media, across reboots of the VMs. The test code we wrote for 3.7 will (A) send a request to interface X endpoint requesting the launch of a Linux VM with some customization (identify images, CPU cores, memory, network and/or firewall settings), and (B) after verifying the VM is up and running, will connect to the VM (using a Java SSH2 client library using pub/private key authentication) to run some simple Linux command to show the subscriber has secure "launching and administration" of the VM. This style of testing has been implemented and run against open source implementations we had internally for EC2 interface, public commercial EC2 service interfaces and now we are doing it against OCCI interface exposed through OpenNebula 2.2 installation we had internally. I understand the OCCI interface through OpenNebula 2.2 is not the latest OCCI spec, however, I believe the semantics it exposes (and the protocol patterns) are not too much different from OCCI 1.1. Of course, if we can get to an endpoint (either install locally or made available to use on the internet) that has OCCI 1.1 interface and can demonstrate the feasibility of the above requirements, we'd be happy to update the testing code and run it to reflect that. A larger goal for the testing results is to let the cloud consumer know that their requirements can be met here and now by interfaces X, Y and Z. Hope this helps clarify what we do on SAJACC and the relevance of testing against a real interface endpoint. Thanks, --Jin Jin Tong NIST Cloud Computing Program Associate [1]: SAJACC Cloud Computing Use Cases, draft: http://www.nist.gov/itl/cloud/use-cases.cfm [2]: SAJACC collaboration twiki site: http://collaborate.nist.gov/twiki-cloud-computing/bin/view/CloudComputing/SA... -----Original Message----- From: Alan Sill [mailto:alan.sill@ttu.edu] Sent: Tuesday, May 03, 2011 2:25 PM To: Tong, Jin Cc: occi-wg@ogf.org Subject: Fwd: [occi-wg] SAJACCTeleConf13 < CloudComputing < TWiki Jin, can you describe the source of your test code calls? perhaps we can get better participation from the occi-wg here in getting a true OCCI v1.1 implementation test going. Alan Begin forwarded message:
From: Andy Edmonds <andy.edmonds@gmail.com> Date: May 3, 2011 1:17:06 PM CDT To: "Sill, Alan" <alan.sill@ttu.edu> Cc: "occi-wg@ogf.org" <occi-wg@ogf.org> Subject: Re: [occi-wg] SAJACCTeleConf13 < CloudComputing < TWiki
Having quickly looked at the code I can say this is not OCCI v1.1. In fact it appears to be OpenNebula requests in the code. Many of the current OCCI implementations (e.g. Platform's, ralfs and my own) contain a large amount o unit, integration and functional test - these could be used to help the efforts of SAJACC
On 3 May 2011, at 20:10, Alan Sill <alan.sill@ttu.edu> wrote:
The dial-in information for the weekly meeting is as follows:
* Phone: (Toll Free) 866-819-5964 (Toll) 1-203-277-6836 * Participant code: 5576521 The URL for the web conference tool for the meeting is: http://webconf.soaphub.org/conf/room/sajacc
It is possible and encouraged to participate remotely.
On May 3, 2011, at 1:02 PM, Sill, Alan wrote:
Thought you would be interested in the following link. OCCI is front and center on this call!
Link: http://collaborate.nist.gov/twiki-cloud-computing/bin/view/CloudComputing/SA...
Alan Sill, Ph.D Senior Scientist, High Performance Computing Center Adjunct Professor of Physics TTU
==================================================================== : Alan Sill, Texas Tech University Office: Drane 162, MS 4-1167 : : e-mail: Alan.Sill@ttu.edu ph. 806-742-4350 fax 806-742-4358 : ====================================================================
_______________________________________________ occi-wg mailing list occi-wg@ogf.org http://www.ogf.org/mailman/listinfo/occi-wg
occi-wg mailing list occi-wg@ogf.org http://www.ogf.org/mailman/listinfo/occi-wg
Alan Sill, Ph.D Senior Scientist, High Performance Computing Center Adjunct Professor of Physics TTU ==================================================================== : Alan Sill, Texas Tech University Office: Drane 162, MS 4-1167 : : e-mail: Alan.Sill@ttu.edu ph. 806-742-4350 fax 806-742-4358 : ====================================================================