Your Custom Text Here
Mary Meeker's Internet Trends presentation has become one of the most relevant presentations in the Internet industry.
Below is the presentation. You can walk through it. I did, and I feel good that the stuff I am focusing on aligns with her data. It's great when you can stay ahead of what the analysts present. :-)
There are a fair amount of ex-nuclear sub staff who work in data centers. It is possible the idea of a small nuclear plant could follow at some part far in the future. MIT Review discusses the current state of small nuclear reactors.
Nuclear option:Babcock & Wilcox’s proposed power plant is based on two small modular nuclear reactors.
Small, modular nuclear reactor designs could be relatively cheap to build and safe to operate, and there’s plenty of corporate and government momentum behind a push to develop and license them. But will they be able to offer power cheap enough to compete with natural gas? And will they really help revive the moribund nuclear industry in the United States?
Last year, the U.S. Department of Energy announced that it would provide $452 million in grants to companies developing small modular reactors, provided the companies matched the funds (bringing the total to $900 million). In November it announced the first grant winner—Babcock & Wilcox, a maker of reactors for nuclear ships and submarines—and this month it requested applications for a second round of funding. The program funding is expected to be enough to certify two or three designs.
Natural gas is so cheap it is hard to imagine a small nuclear plant being deployed any time soon for a data center.
Even if small reactors can compete with conventional nuclear power, they still might not be able to compete with natural-gas power plants, especially in the United States, where natural gas is cheap (see “Safer Nuclear Power, at Half the Price”). Their success will depend on how much utilities think they need to hedge against a possible rise in natural-gas prices over the lifetime of a plant—and how much they believe they’ll be required to reduce carbon dioxide emissions.
“At the end of the day, we’ll build the lowest-cost option for ratepayers,” Cryderman says. “If it’s too expensive, we won’t build it.” The challenge, he says, is predicting what the lowest-cost options will be over the decades new plants will operate.
One of these days natural gas will not be plentiful and nuclear is going to be one of those options that may make sense.
The data center is more and more strategic to many businesses. It is now common for outages to cost $10k-100k/min. Many of the data center executives have military backgrounds and are used to defending their country. Some data centers are built like fortresses with even armed guards inside the building. Many times it is not the outsider that brings down a data center, but the insider who makes a mistake in operations and maintenance. These employees though are not the enemy. The enemy that has attacked you is the outage itself. When an outage occurs you can run through a playbook that lists the standard approved operating procedure which is fine if you have the time and the outage scenario was covered in your planning.
What happens when the outage is something that you had not planned for. You run the playbook, can't figure out how to address the outage, and now you are thinking crap. What do we do now? Outages can kill a company or business unit if data is destroyed or downtime is excessive. Think of the T-mobile Sidekick outage.
The incident caused a public loss of confidence in the concept of cloud computing, which had been plagued by a series of outages and data losses in 2009.[7]
Was the enemy the Microsoft employees who ran the services. No. The enemy is a collection of ideas of what was the right thing to do which eventually caused an outage.
A company statement said the mishap was due to "a confluence of errors from a server failure that hurt its main and backup databases supporting Sidekick users."[2] T-Mobile blamed Microsoft for the loss of data.[1]
Someone had the idea to insure the uptime of the Superbowl is to install a protection relay.
“The purpose of it was to provide a newer, more advanced type of protection for the Superdome,” Dennis Dawsey, an executive with Entergy Corp., told members of the City Council. Entergy is the parent company of Entergy New Orleans.
Entergy officials said the relay functioned with no problems during January’s Sugar Bowl and other earlier events. It has been removed and will be replaced.
4 years ago I read about John Boyd and his OODA Loop approach and posted on it. I tried finding more details on what John Boyd presented. His presentations are difficult to understand and unfortunately John Boyd did not write his ideas down well enough for others to understand. Then I found a PhD thesis by a military student who did take the time to explain John Boyd's ideas. You can find it here. Warning this paper is for people who really want to understand modern military strategy. The OODA loop concept has been transferred to business on the idea of the winners are those who can move faster and out think their opponent.
Who is John Boyd?
a tribute written two days
after Boyd’s death on 9 March 1997 which describes him as
a towering intellect who made unsurpassed contributions to the American art of war. Indeed,
he was one of the central architects in the reform of military thought which swept the
services, and in particular the Marine corps, in the 1980’s. From John Boyd we learned about
the competitive decision making on the battlefield-compressing time, using time as an ally.
Thousands of officers in all or services knew John Boyd by his work on what was to be
known as the Boyd Cycle or OODA loop. His writings and his lectures had a fundamental
impact on the curriculum of virtually every professional military education program in the
United States-and many abroad [..]he was the quintessential soldier-scholar - a man whose
jovial outgoing exterior belied the vastness of his knowledge and the power of his intellect1.
The problem the author, Frans Osinga was trying to address was the lack of explanation of how Boyd came to his conclusions. What was his logic and assumptions?
There are a number of short papers35. Most if not all deal almost exclusively with the
OODA loop concept. Recently, two biographies have appeared. Robert Coram’s work
focuses in particular on Boyd’s life and less on Boyd’s strategic theory, although he does
provide a good synopsis of it. Boyd’s biographer Grant Hammond surpasses Coram in his
rendering of Boyd’s strategic theory but the book nevertheless falls short of offering a
comprehensive account of Boyd’s work. Instead it must be considered an authoritive and
very accessible description of Boyd’s ideas. Moreover, as it does not contain an integral
rendering of Boyd’s work, the educational experience contained within Boyd’s slides, his
unique use of words and the way he structures his arguments, does not receive the emphasis
it deserves. Finally, although touching upon Boyd’s wide array of sources underlying his
work, space restrictions prevented a proper discussion of the intellectual background of
Boyd’s work.
I am slowly digesting the PhD paper. You can also buy the PhD paper in a book.
This book aims to redress this state of affairs and re-examines John Boyd’s original contribution to strategic theory. By highlighting diverse sources that shaped Boyd’s thinking, and by offering a comprehensive overview of Boyd’s work, this volume demonstrates that the common interpretation of the meaning of Boyd’s OODA loop concept is incomplete. It also shows that Boyd’s work is much more comprehensive, richer and deeper than is generally thought. With his ideas featuring in the literature on Network Centric Warfare, a key element of the US and NATO’s so-called ‘military transformation’ programmes, as well as in the debate on Fourth Generation Warfare, Boyd continues to exert a strong influence on Western military thinking. Dr Osinga demonstrates how Boyd’s work can helps us to understand the new strategic threats in the post- 9/11 world, and establishes why John Boyd should be regarded as one of the most important (post)modern strategic theorists.
I am still thinking of the 75% 55-65 audience of Uptime's event, and whether understand how big and fast the shift to mobile is coming. The trouble about most data center building type of people is they have no idea what kind of hardware, let alone software is running in the data center.
Now you may think what's the big deal. KPCB's Mary Meeker just shared her 12/3/2012 Internet trends and in less than three days there are 400k + views of this presentation.
This is the slide that illustrates the change.
There are 88 slides in the deck.
If you think Mobile means iOS, wake up the rapid growth of Android. I have iOS and Android devices. Do you?
With little bitty screens, the mobile devices still are pulling down 13% of the traffic.
And India now has more mobile traffic than desktop.
I made the strategic shift to Mobile a while ago. Have you?
Along with all this mobile growth is the growth of Big Data. Mobile + Big Data is the new frontier.
And part of this shift is the transition from an Asset Heavy lifestyle.
To Asset Light. Which works really well in high growth emerging markets.
Which turn off many of you.
But, keep in mind. This asset light approach is driving new flexibilities.
GigaOm's Stacey Higginbotham posted on Spanner: Google’s Globally-Distributed Database.
Google’s Spanner: A database that knows what time it isBY Stacey Higginbotham
Google, which is notoriously secretive about technology advances, has opened up the vault and spit out a research paper on its Spanner database. And like other Google innovations, this one is hot. It’s a database that scales to millions of machines and trillions of rows.
The Research Publication is published here.
Abstract
Spanner is Google's scalable, multi-version, globally-distributed, and synchronously-replicated database. It is the first system to distribute data at global scale and support externally-consistent distributed transactions. This paper describes how Spanner is structured, its feature set, the rationale underlying various design decisions, and a novel time API that exposes clock uncertainty. This API and its implementation are critical to supporting external consistency and a variety of powerful features: non-blocking reads in the past, lock-free read-only transactions, and atomic schema changes, across all of Spanner.
To appear in:
OSDI'12: Tenth Symposium on Operating System Design and Implementation, Hollywood, CA, October, 2012.Download: PDF Version
I have got a bit more time to go through the paper and I'll highlight parts I find interesting from a data center analyst perspective.
First thing caught my eye.
Applications can use Spanner for high availability,
even in the face of wide-area natural disasters, by replicating their data within or even across continents. Our
initial customer was F1 [35], a rewrite of Google’s advertising backend. F1 uses five replicas spread across
the United States.
What is F1?
And how does this relate to Spanner?
Google has 3 to 5 data centers in a geographic region.
Why 5, not just 3?
I had just written about Google's move to Chile had three more. So, need to get used to looking for 5 Google data centers to support a geographic reason with 100ms latency, not just three.
I could go longer into the document, but I think the post will get too long. Let's just stop with one point that is useful in looking at these two documents. Google has 5 data centers within 100ms in a geographic region to support ad business which is where they make most of their money.
