"Open source cloud platform is commercialized by its creators" @
TL
I'm impressed a pre-revenue company managed to raise $5.5m basically implementing others software & API's. One of the reasons we haven't implemented the EC2 API for our Enomaly ECP isn't because it's difficult but because we have no assurances from Amazon that they won't sue us for our implementation. Now that Eucalyptus has gone legit, what's stopping Amazon from sending a cease and desist? If amazon publicly said "you are free to implement the EC2 API within your platform" with a clear picture of their related patents, we'd have it ready in a few weeks. Problem is most of my conversations with Amazon have indicated otherwise.
My other issue with basing our API on EC2 without a public roadmap is we're going to be always a few steps behind scrambling everytime there is a new feature released. Working against our own roadmap is hard enough.
Hi Harald,
Your question is a good and very important one. We have spent some time with the UCSB legal department trying to make sure we aren't violating anyone's IP. More importantly, it is definitely not our intention to harm Amazon's interests in any way. In fact, our intention is quite the opposite. We hope that EC2 will become even more popular based on experience with Eucalyptus. With that said, Universities are extremely resource constrained in this regard (and Amazon is not) so we may get squashed in the end.
However, here is our (perhaps imperfect) understanding. There are three areas if IP in play: copyright, trademark, and patent. In terms of copyright, we have not directly incorporated anything from Amazon into Eucalyptus (not even the WSDL). Amazon makes its WSDL publicly available and it does not appear to be copyrighted (at least, we couldn't find anything even hinting that it was copyrighted) but even so, we do not include it -- not even for the purpose of generating the code for the web services front end. However it is absolutely true that the hand-written code for the front-end was written with an understanding of the WSDL. Turns out that the internals of the Eucalyptus front end are set up to support more than just the EC2-generated messages so there is not a completely direct mapping but if it does turn out that the WSDL is copyrighted, there may be an issue here with Eucalyptus as a derivative.
As far as trademark goes, we can't find anything that indicates trademarks associated with the function names or message names. We didn't spend a lot of time trying to obfuscate the internals so there may some overlap. If that is a sticking point, we will certainly rename everything.
Finally -- patents. Again, we did some due diligence and we couldn't find anything to indicate that what EC2 is doing is patented. There are parts of S3 that are patented (notice that we don't have S3 support in Eucalyptus) but it may also be that the patents are pending so our search didn't pick them up. If they are, Eucalyptus (and pretty much every other cloud computing infrastructure, I suspect) will need a license from Amazon.
Cheers,
Rich
@TL Agreed. $5.5M can solve a lot of probs, for sure....but as Ruv says, they're dancing with the elephant and they could easily get squished....
As for the API issues, I think that's FUD. Sure, they could copyright the WSDL, but this is no different than dozens of 'fair use' and reverse engineering efforts that have been successful in mimic observed behavior. Samba, Open Office leap to mind. Many others for sure.
As for this 'patent', not sure I'd run scared of a application that's likely to get rejected, or at least narrowed in scope before anything is granted. [If anyone wants to buy me a beer some time, I can tell them about the millions I spent on patent litigation. The system is totally F'd']
They've got to have more than EC2 compatibility since, as Ruv says, thats not rocket science.
Regardless, I think EC2 compatability is simply a tactic (or it ought to be). Their sights are proabably someplace else, like VMware. And that, I'm afraid, is a lot more like rocket science and $5.5M is only going to get them part of the way there.
The Eucalyptus license also makes mention of patents in a relatively
non-standard way (I.E. it is yet another BSD variant of its own). I
wouldn't immediately jump to believing it compatible with other
licenses so easily.
As far as the GPLv3 is concerned, I know that not everyone likes the
GPL, but there are many that would see this as a plus. Personally,
I'd prefer the AGPL.
Regards,
Eric Windisch
I agree with @Reuven that implementing the Amazon APIs is a dangerous game. Today Eucalyptus is a nice company, but if they grow they won't be so nice. If Amazon feels that they can be a threat, then Eucalyptus could have not only legal problems, but product roadmap problems. So sooner or later Eucalyptus will not use the AWS compatibility as a key feature, to avoid 'AWS API lock-in'.
It's worth noting that a good deal of the publicity Eucalyptus received in the lead up to the funding was due to their support of the EC2 APIs so it may well be that this was nothing more than a (clever) marketing ploy...In our view, Eucalyptus respects Amazon's IP in two important ways. First, it is not a "reverse engineered" version of AWS as it is not designed to operate on the same scale that AWS is. Put another way, we designed Eucalyptus to run in a typical academic cluster environment and not across several massively provisioned, geographically distributed data centers (as is AWS). If that scale were a feature of the target setting for Eucalyptus, we would have architected it very differently. Because it is not, we did not consider or try to deduce how AWS might be engineered as it must certainly take into account contingencies at frequencies that simply do not occur at the cluster scale.
Secondly, it has always been the intention of the project to provide a vehicle for stimulating cloud usage and experimentation in general, and AWS usage and experimentation in particular.
For example, we have encountered scientists and academics who would like to use AWS, but who must justify the expense (however paltry) to their funding agencies using local resources. These research groups have application code bases in which hundreds of person-years have been invested (read: several generations of graduate students and postdocs have authored the code). Eucalyptus can be used to demonstrate that these codes will work "well" (or least, will work at all) once transitioned to AWS.
Thus we have proceeded under the assumption that we have not infringed on Amazon's IP with respect to AWS internal functionality and that we have designed a "tool" that facilitates greater AWS usage. It has never been our intention to harm Amazon's interests in doing so.
2009/5/5 Diego Parrilla Santamaría <diego.parril...@gmail.com>I would not consider this a reason to avoid Eucalyptus - they have been well aware of the risk since the outset and claim to have designed their system to cater for multiple APIs... I'd obviously love to see them implement OCCI (which could well become the HTTP of Cloud Computing) when it's available later this month but whatever happens I'm expecting to see one or more new APIs on their roadmap before long if they're not already. This more recent response is insightful too:I agree with @Reuven that implementing the Amazon APIs is a dangerous game. Today Eucalyptus is a nice company, but if they grow they won't be so nice. If Amazon feels that they can be a threat, then Eucalyptus could have not only legal problems, but product roadmap problems. So sooner or later Eucalyptus will not use the AWS compatibility as a key feature, to avoid 'AWS API lock-in'.