 Welcome, everyone. This is the second question and answer session for the ARDC Nectar Research Cloud Refresh RFP. I'll just very briefly. ARDC has some funding to provide a refresh to the existing Nectar Research Cloud with the idea of maintaining the capacity of what is now quite old equipment. So, at the moment, the RFP is out in a draft form for feedback and for your information. The final version of the RFP will be released on the 24th of October and then there will be a two-week period for responses. And then the decisions will be made by the end of November. So, it's a fairly short time period for responses because you will be able to have seen the RFP for several weeks beforehand to put ideas for proposals together and to discuss them with ARDC if required. So, as part of the RFP process, we've got the draft RFP out there and we wanted to have a couple of Q&A sessions where anyone can ask any questions about the RFP process. Any questions in this session will be and the answers will go into the FAQ so that anyone can have a look at the answers we've provided to those questions. So, unless Ian, is there anything else you want to say as an introduction? No. So, we'll just kick it off and get started if anyone has any questions, far away. I thought you might Steve. Yes, you can be first. So, part A, page 13 talks about co-investment in section 4.4. One of the items for co-investment is infrastructure maintenance. During sort of when you do your acquisition, typically you will acquire the equipment with the number of years maintenance sort of as part of that initial purchase. Is that maintenance component need to be separated out as co-investment or is that just part of the initial capital purchase? That's a good question. I don't think that's something we thought about. Ian, do you have any views on that? Yeah, it is a good question. I think if the co-investment is going to be regarded as if the maintenance is going to be looked at as part of our investment, it's not a piece that we would pay for out of the capital funding. So, let's have a little bit of a think about how we deal with that. It is a fairly standard thing. Not everyone tends to put things under maintenance, but I suppose most people would do. I would imagine possibly, and again I'll discuss this with Ian and we'll come up with an answer we'll put on the FAQ. I would have thought if you wanted to have it as part of what was funded that would fine, but you'd have to match it with additional co-investment. If you wanted to have it included as something you pay for that counts towards your co-investment, then that would also be fine. So, I think to some extent we might just leave it up to you as to how you would want to play that. But I'll confirm, we'll have a chat afterwards with Ian and we'll confirm that and put in the final answer in the Q&A. Okay. The FAQ, sorry. Yeah. Next question is on page 14, which is with respect to practical completion. So, you need to operate this for three years. So, is practical completion once it is commissioned or at the end of three years? Yes, it's once it's commissioned. I'll check if that's not clear in the document. But yeah, the idea is the infrastructure needs to pass some acceptance tests. We have to agree that it's ready, it's up and running, and at that point we'll sign off that you're done, essentially, because the capital funding is really just to the point where or the payment is just to the point where you have the infrastructure up and running. So, it's a final payment milestone. So, yes, that's the completion of the project, essentially, where we fund, at least in terms of the capital part. Yeah. Part B, page four. Let's get to that. So, there's a reference to an ARDC dataset and collection node. Is that just the old RDS nodes? Yes. That's the term for that. Okay. Page five, 1.3. So, organizations provided funding under this program, research cloud nodes. And in section four, it talks about the steering committee will include representation from the successful research cloud node provided. So, this was just that comment, Paul, just to confirm. So, if an existing node is unsuccessful as part of this process, are they still part of the steering committee, or is it only the successful nodes under this program? Yeah. We haven't confirmed what the new steering committee will look like, but the expectation is it would be the major stakeholders. So, I would expect it wouldn't just be the successful groups if there are existing nodes that want to continue to federate into the research cloud, then they, I expect, would be part of the steering committee as well. It's really, to a large extent, the point of that steering committee is to acknowledge that this isn't just an ARDC service, right? It's an ARDC service done in collaboration with our partners, and those partners need to have a say in how that service is run. So, that would include previous nodes if they are still part of that service, yes. Yeah. Is there any section four also talks about the Technical Advisory Committee and the Resource Allocation Committee? Is there any indication as to the makeup of those? Does each node have a member of the TAC, or is it just sort of the senior technical people? So, again, we haven't decided on the makeup of that. And to some extent, that would, I think, partly be directed by the governance sort of steering committee group. But I would expect that it would include representation from, again, the partners in delivering the cloud infrastructure and possibly some external experts as well. Part D, page nine. Good to that. So, it's around needing to fill out for the research sector needs and benefits for seven, eight and nine, given this is effectively a refresh of previous or existing infrastructure, do we need to go through the completing out target research community? We're not looking at significant, you know, we're not looking for you to have to put in a huge amount of, you know, listing all the research groups you support or whatever. It's really just as it was in the initial phase, sorry, the initial Nectar RFP, just trying to give an indication of the major research communities that you are looking to support. So we're not, you know, it's not meant to be an exhaustive list by any means. It's just trying to give an indication of, you know, some of the groups that are important, you're supporting. And so I think it is useful to put some of that in, given that, you know, this refresh is particularly focused on supporting national, nationally prioritised research and, you know, increased capabilities and so on. It's important to at least put something in there to address that issue. Yeah. And page 11, with respect to number 18 around communications and engagement, it says described the means by which customer satisfaction with the proposed projects planning, etc will be measured. I wasn't around when the initial one was produced. And like I know we run a customer satisfaction survey annually, but that's more around, you know, the outputs of and use of the infrastructure and just curious as to whether there's any advice around how you would measure customer satisfactions with project planning. Yeah, that part, again, was carried over from the original one. And I suppose the point there was just to try to ensure that the people you're, you know, the main groups I was just talking about, there's sort of main national groups that you're trying to support here are essentially part of the discussion of the sorts of equipment that you buy. We don't want to get to the point where, you know, we roll out all this equipment and the key groups we're trying to support come and say, well, that doesn't really meet our needs, right? It's the wrong, it's the wrong sort of stuff. It's not enough or whatever. So it's really just to clarify how you are, I guess, give some confirmation that you are allowing those groups to provide some input into, you know, what sort of infrastructure and what size of infrastructure is actually is going to meet their requirements in fact. Yep. Okay. So Section 2.2 talked about this acquisition was predominantly towards supporting the standard flavours. So if the research communities that we're supporting are looking for things beyond that, it's acceptable to effectively have a proposal that just supports the, say, a large memory node or large memory nodes and GPU nodes as opposed to necessarily the predominant flavours. I don't know if it would be acceptable to say it's just that and nothing else. I would have thought, I mean, it does say in the RFP, we're expecting to mostly support the standard flavours, but if you have a good reason with requirements from research groups that they want, big memory nodes or GPUs or whatever, you should certainly say that and then that's fine. As long as there's demand and that meets requirements for nationally prioritised research groups, having some of that sort of non-standard infrastructure is fine. I wouldn't have thought that would be, you know, taking to account the entirety of the groups that we want to support with national prioritised research, the only thing they would want would be GPU nodes and big memory nodes. I would have thought there would have to be at least some capacity from the standard flavours. But in terms of how much non-standard stuff you want, yeah, if you can make a case for it, then that's fine. Alright, thank you. They're my questions. Thank you. Great. Thanks. Very good questions. Thanks, Stephen. Ian, sorry, was there anything I said there that you wanted to elaborate upon? I think just touching on that last question, it's worth remembering this is a capacity maintenance. It's not this stage of the investment in the cloud is not so much around evolving it into high memory nodes or slightly different flavoured cloud. So just bear in mind those non-standard flavours, there needs to be some pretty robust discussion around that and how that might impact on the people who are relying on the more standard flavours at the moment. Thanks, Ian. Okay. Do we have any other questions? Leslie, I can't believe you're not asking a question. Okay, anyone else want to put up their hand for something? No, it could be a short session then. Is it worth pointing out, Paul, that we're going to try and actually fit in another draft and another consultation session after this. So we do have time to do that again. We're also trying to make sure that the drafts as they appear highlight the changes, so the deltas. Also, I think there's a word copy of the documents up there now, so people can start filling it out and seeing where they run into some of these questions. Yeah, so now in the documents, as well as the second version of RFP, there is a document there on the website now that specifies what the changes are from version one, which are all fairly minor and a word version, as Ian said, if you want to start filling it out. So what we're looking at is there's still a couple of things in the RFP, excuse me, that aren't quite specified yet, and that includes sort of required specifications, the details that required specifications for a node to meet to include into the Federation. So we have an early draft of that, we expect to have that finished in a couple of weeks. And we're expecting essentially to have an X version of the RFP that basically is done or pretty much has been done. So we're looking, plus a draft contract for people to look at well beforehand and flag and issues and so on. So that's we're expecting to have that out by the end of the month. So essentially, that would be a version three of the RFP and a draft contract and all the stuff that's currently, you know, a couple of things that are currently missing from the current RFP, the cloud specifications, the acceptance testing specifications. So that should all be done by the end of September. And we wouldn't expect, we'd expect very little change from the documents from then until they're released in the 24th of October. So you should be able to put your proposal together well in advance of that, using those documents. One further question, Paul. So both Dell and HP have got where we're coming up with some high density options with respect to compute nodes around the Epic Rooms. And we've seen sort of one presentation from HP expecting to see something from Dell shortly, but the pricing on those might still be a little way off before we can see those are availability for those nodes. So is that this stage, if we hit November and we still haven't seen sort of firm pricing, we're just looking at an indicative thing at this point and clarify that post the proposal? Yeah, I think so. You could put indicative pricing or you could put, I mean, standard thing that we tend to do with HPC leaf grants in the past, you put something in that's based on existing equipment and pricing for existing equipment. And then, you know, a couple months later, when it's all good to go, you might revise that based on latest pricing and whatever. So I think that the plan for the RFP is you basically put in a fairly detailed specification of and costing for what you want in the RFP. It's going to take, you know, a month for us to decide on that. It's going to take a while to contract it. And then the next phase is you would put in a essentially a quote to say this is exactly what we're going to buy and we and Iodice would approve it. So that's not going to happen before January. So I thought there's certainly scope for you to revise what you'd propose based on new pricing, etc, etc. And as long as it doesn't deviate too much from what was agreed, I would have thought we would say that's fine. Okay, thanks. Ian, does that seem reasonable? Yeah. All right. So one final chance for anyone to ask any final questions before we wrap up. No. All right, I think unless anyone Ian anyone else has any comments before we go, then I believe we're done. So as I said, the questions that were asked today will will be added to the FAQ. And this video will be put up on on YouTube for others to have a look at who weren't able to make it. Right. Thanks very much, everyone for attending and have a good morning. Thanks for