Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : The *NEW* Frank Coluccio Technology Forum -- Ignore unavailable to you. Want to Upgrade?

To: elmatador who wrote (46179)	1/20/2018 7:56:19 PM
From: Frank A. Coluccio	Respond to of 46821

Great find, Elmat, thanks. Throughout the years, on this and other fora, I have been speculating about the shape of the next pendulum landing zone, to the point that I had t. Frequently, I came up short, motivated by a need to reduce latency, I stopped suddenly at the end points where a kind of ambient storage/processing would take place. I just didn't equate it with IoT, per se. Here's an interesting exchange from about a year ago, started by forum member elsewhere, wherein you and I, engineer, and others chimed in with similar thoughts: siliconinvestor.com FAC ------

To: elmatador who wrote (46179)	1/22/2018 10:18:51 PM
From: Glenn Petersen	Read Replies (2) \| Respond to of 46821

Backchannel (now part of Wired) had a good article on Edge Computing last August: Message 31241376

To: elmatador who wrote (46179)	1/31/2018 2:09:00 AM
From: elmatador	Read Replies (1) \| Respond to of 46821

After The Economist picked my piece on the change of datacenter geography. I wrote that the Chinese will take over African connectivity Will China takeover African connectivity? The Harvard Political review took it over from there Internet, Inc.: Technology Superpowers and the Future of African Connectivity harvardpolitics.com

To: elmatador who wrote (46179)	2/9/2018 9:00:37 AM
From: Frank A. Coluccio	Respond to of 46821

Hi Elmat, Further to our upstream discussion on the shifts taking place in data center architectures, I posted the following article to the Cook Report Architecture-Econonics Discussion Group earlier today in response to someone's question concerning if economics was still a topic of interest there: Some of us look at the economics closely, some very closely. It's that we don't discuss it very much anymore, and I'm guessing this is due to a prevailing notion held by members similar to that which you just expressed (above). So thanks for breaking the ice ;) Until I read your comment I was on the fence about posting an article on the economies and dis-economies of scale in a continuum of data center sizes and designs; telecoms networks employing GPON technologies; and, (drum roll) in the design of space vehicles, as well. In the telecoms part, I spoke with a colleague about how economies (and dis-economies) were achieved, illustrating how break-even benchmarks are established and exceeded, and how modulating the number of end points downward from the break-even point (as well as a host of environmental variables) can result in insufficient scale of operations, manifesting increasing the dis-economies of scale, proportionately. The conversation was in some ways similar, and in some ways diametrically opposite, two articles I'd came across earlier in the day. The first article was on the scaling (and physical sizing) of server populations within hyper-scale data centers, in contrast with the second article that discussed the number of booster engines now being employed in SpaceX rockets. While the former seems to have reached a tipping point for certain compute situations which have reached massively hyper-scale proportions, argues Amazon's AWS data center architect James Hamilton (see article, immediately below), the latter (rocket designs) are now in the early phases of scaling up the number of parallel engines used, as explained by Elon Musk in today's ars technica article (bottom). *Tensor Processing Unit* By James Hamilton \| *Perspectives* blog \| Amazon AWS perspectives.mvdirona.com *Musk explains why SpaceX prefers clusters of small engines* By Eric Berger \| *ars technica* arstechnica.com I'd be interested in reading any and all comments, corrections, etc. FAC ------