Yes that will be it. They don't own an undersea cable. They could have capacity on SX, TGA or Hawaiki cable systems. All have a path between NZ and AU.
Oops sorry, I didn't add the quote from when Snoopy mentioned this a few posts back :)
Printable View
Yes that will be it. They don't own an undersea cable. They could have capacity on SX, TGA or Hawaiki cable systems. All have a path between NZ and AU.
Oops sorry, I didn't add the quote from when Snoopy mentioned this a few posts back :)
Are these data centres power hungry and dependent on surety of supply?
Just that 0z seemed exposed to power generation shortfall from not enough gas production at the moment and NZ power supply always seems at risk.
Is this a risk to IFT operations?
Or do they just build their own generators and become self reliant?
Slide 25 is showing a total shareholder return of 18.7% per year since inception. As the company gets bigger the same percentage increase per year gets harder to achieve. Yet over the last five years, the company has achieved a total shareholder return of 24.6% per annum, which is incredible. The slight downside to this is that most of these returns have been from independent valuations of existing business units, rather than sales 'at market'. And valuations have assumptions, like growth rate discount rate etc. etc. . I am not wanting to cast aspersions on the valuers here, who I am sure have done a thoroughly professional job. But to better understand this valuation process, the latest annual report makes interesting reading
From AR2024 p34:
"Infratil’s investment in CDC is now valued between A$3.8 billion to A$4.4 billion, up from A$3.1 billion to A$3.7 billion 12 months earlier. This valuation increase reflects the significant additional demand and future development pipeline added over the course of the year, somewhat offset by an increase in the required return to take account of the increased risk profile associated with developing at scale. The independent valuation assumes 1,220MW of total built capacity, up from 585MW 12 months earlier. This
consists of 268MW operating capacity, 416MW under construction and 536MW classified as future builds."
I read that as the total current pipeline of data centre construction for some years into the future, being already priced into today's returns. I thought it interesting that the valuers even admit that their valuation is based on the 'build it and customers will come' business scenario, (when talking about the extra risks of developing at scale).
So am I saying that this 'return' Infratil shareholders have had these last few years is all BS? No, I am not saying that. But I do question the ability of valuers to realistically value a data centre portfolio when the total data centre capacity from all market players to come up for commissioning in the near term will increase total market data centre capacity by 500%.
SNOOPY
I think with hindsight we can forgive the European Airport fiasco. Infratil was coming off a very rewarding period of investing in Wellington Airport, which I recall some 20-25 years ago was known for having more flight movements than Auckland (albeit not more passengers as on average the planes using Wellington Airport were smaller.) With the growth in international tourism, I am not sure if that is still true today. But Wellington remains a 'busy little airport' nevertheless.
So I can see where the interest of investing in other 'secondary airports' came from. But as it turned out, those European Airports that Infratil chose to invest in did not have the same 'essential hub' characteristics that Wellington Airport has. We pay business people to take risks, and sometimes those risks do not pay off. That's business. And as long as management learn from their mistakes, I would be prepared to forgive them.
SNOOPY
Definitely Forgiven - but not forgotten! The global credit crunch did not help passenger numbers.
I think Ardmore has more plane and helicopter flight movements than Mangere too. Planes smaller and older though. A fair amount of new building there too. It does help to have pivotal or unique characteristics.
As the above reference reads. Projecting an incremental 500MW data centre power load from the generators over the next five years. The equivalent of 200,000 more houses in Auckland.
Yes
You would have to think a data centre operator would have a back up generator supply on site wouldn't you? Otherwise the whole case for being a 'reliable' data provider goes out the window. But upwards of 20MW as a back up generator on site sounds pretty big. I wonder how data centre providers do manage a 'power out' scenario? Anyone know?
SNOOPY
Let google enlighten
https://kenshaw.com.au/case-study/ca...ntre-cdc-2018/
From the link
"CDC has also commissioned Kenshaw to install 4 x 2750kVA Cummins Generators with custom canopies in their brand-new Data centre due for completion in late 2018."
I wonder how big that datacentre in Australia was when it opened in late 2018? The reason I ask is that a 2750kVA Cummins Generator, might otherwise be reported as a 2.75MW Cummins generator. Four of those would produce 4x 2.75MW= 11MW.
That is not enough power to keep a more typically sized for today 20MW data centre on line. CDC would need 8 x 2750kVA Cummins Generators to support that. Crikey! I might have to buy some shares in Cummins!
SNOOPY
I'd say there would be a combination of UPS to keep the important things on for a period of time(and provide a clean power supply) as well as some smart load shedding to drop off the unimportant things.
Ultimately all data centres will need to go to where the power is cheap and reliable, even if it is remote. If not remote, it makes sense for HPC to be colocated with rapid (15 second for example, or faster) curtailment activities such as Bitcoin mining. The overall HPC demand will be more than most people can even imagine, and that's with HPC technology advancement over time. It might take a boom and consolidation to sort this out just like the internet. Yes datacentres have backup which is typically diesel generators onsite but as battery technology improves over time we will see this gradually migrate to battery and with combination HPC-mining you have the best of all worlds.
From my AI app:
### Conclusion
While a power loss at a CDC Data Centre can have significant implications, the robust infrastructure and redundancy measures in place are designed to mitigate impact and ensure that services remain operational. These measures include UPS systems, backup generators, redundant power feeds, data replication, and comprehensive incident response plans. However, the efficiency of these systems can vary, and there may still be brief service interruptions or impacts depending on the specific circumstances and the effectiveness of the data center's contingency planning