What happens when the Internet crashes?

http://docandersen.podbean.com
https://docandersen.wordpress.com
http://scottoandersen.wordpress.com
My Amazon author page!!!!
http://lukeoandersen.wordpress.com
http://chuckandersen.wordpress.com
http://NickOandersen.wordpress.com
http://content.iasahome.org/blog
http://www.safegov.org

I am worried. I published a blog on http://www.cloudtweaks.com on the total bandwidth available and could that bandwidth support 100% of the world’s population consuming Internet services.

The answer is no.

The problem is growing. First off the number of things that are on the Internet and available for consumption is growing. From Netflix and Hulu to the various network services all are available. You can watch your DISH, Direct TV, Comcast, Verizon and other TV providers content anywhere on any device (well not Windows 8 phones in all cases).

You can consume your corporate email from home.

The speed of light is more than an Albert sat down and figured out that light has limits, it is the current maximum. Yes it is conceivable that in fact humans will break that barrier in the future (opening up 10 times 10 to the 10th more bandwidth) but until then there is a cap on the total available bandwidth.

CDN solutions (Akamai, Level 3 etc.) provide an interesting partial answer to the problem as do WAN Accelerators (Riverbed). They reduce the duplicate packets being sent around the network. But at most going forward for the entirety of the Internet that is going to be a 10% savings overall. That as much as to do with the uniqueness of data requests going to different end points. IE, if person a on an iPhone and person b on a android phone both request a movie from Netflix and it happens to be the same movie, it is only common data to the point of separation. IE, if the first person is at Starbucks and the second person is at the office, the point of separation is well before either the companies network for person b, and the Starbucks network or AT&T network for person a.

At some point we will reach the reality of saturation. What happens then? I’ve argued for some time now that for an initial saturation point the Internet will simply crash. Topple over. Reach as Malcolm Gladwell so eloquently wrote “its tipping point.” What happens after that first crash?

.doc

Scott Andersen

IASA Fellow

A little clarity on my blogs…

http://docandersen.podbean.com
https://docandersen.wordpress.com
http://scottoandersen.wordpress.com
My Amazon author page!!!!
http://lukeoandersen.wordpress.com
http://chuckandersen.wordpress.com
http://NickOandersen.wordpress.com
http://blog.iasaglobal.org
http://www.safegov.org

http://:www.cloudtweaks.com

I was added as a blogger on CloudTweaks this week – additional exposure and a chance to share with a different audience. I am now blogging on my blog and on three other blogs. Based on the number of blogs I thought I would share my goals for each.

  • CloudTweaks: Cloud knowledge, information on the concepts and processes of cloud computing. I will aim for once a quarter for these but no promises.
  • IASA: Focused more on in-depth cloud computing and occasionally adding to the What Architects Should column.
  • SafeGov: more focused on government and cloud computing this blog will also have a strong security flavor.

That leaves my two other blogs that are daily. The focus for those will stay similar but I am evolving them a little.

  • DocAndersen (this blog) will continue to focus on technology, architecture and the concepts of innovation. I will add (and did in fact this past year) more serial stories on this blog.
  • ScottoAndersen: This is my personal blog and is more focused on a journal like experience (what is going on around me right now). If you are interested in hearing about my family this is the blog you want to follow.

Its been a lot of fun these now 8 years I’ve had this blog. Thank you to everyone one who has read these pages over the years.

.doc

Scott Andersen

IASA Fellow

Breaking the migration triangle into its supporting components…

http://docandersen.podbean.com
https://docandersen.wordpress.com
http://scottoandersen.wordpress.com
My Amazon author page!!!!
http://lukeoandersen.wordpress.com
http://chuckandersen.wordpress.com
http://NickOandersen.wordpress.com
http://content.iasahome.org/blog
http://www.safegov.org

Considering the concept further of the migration triangle I’ve realized that each of the supporting end points of the triangle has its own triangle underneath it.

The concept of planning for example is comprised of three distinct components:

  • Requirements gathering
  • Technology mapping
  • Risk discovery and documentation

The concept of communications can also be broken into three distinct components:

  • End User Communication
  • Training and helpdesk communication
  • Executive (upward) level communication

Finally the concept of people can also be broken into three distinct components for your project:

  • The people on the project with the skills you need
  • The people you are doing the project for
  • The people reviewing and monitoring the outcome of the project

Of course you could go even more finite and break the smaller buckets into even more smaller components. Such as people who monitor the project not as business leaders but as the next people in the queue with a migration project.

In the end the supporting components are critical. In the people spectrum if you miss one of the three your project runs a serious derailment risk. In planning if you miss requirements or risks your project can be destroyed. Finally if you miss communicating upward and outward you run the risk of being successful with no one ever noticing.

I would argue that in fact the best possible future state for migrations is that they happen and nobody notices. I don’t think that is possible in the near term but that is the dream state for migrations.

.doc

Scott Andersen

IASA Fellow

Migration Triangle continued..

http://docandersen.podbean.com
https://docandersen.wordpress.com
http://scottoandersen.wordpress.com
My Amazon author page!!!!
http://lukeoandersen.wordpress.com
http://chuckandersen.wordpress.com
http://NickOandersen.wordpress.com
http://content.iasahome.org/blog
http://www.safegov.org

Migration Triangle Rules (MTR) continued

MTR 4: The Conops of migration has three parts (the triangle) the first one to fail results in the migration not being completed. Stack enough failed migrations up and you suddenly have a pre-migration wall you have to leap. If the migration fails after some level of completion document the failure so the next team doesn’t have to leap over your failure.

MTR 5: There is a path that many organizations have that has been beaten into well, a path. It leads away from the communications plan once the migration starts. Plan the communications and then act on the plan.

MTR 6: This comes from my days as a Microsoft Solutions Framework training (MSF). No project on earth has ever slipped 6 months in a day. Communication (see above) helps us remind people that projects have delays.\

MTR 7: Migration is a team sport. That means people who start with the concept that they are the smartest person in the room can’t really help you. You may consult them on specific problems but once the migration is rolling get them out of the room.

There are hundreds more rules – I could go on forever. The concept of a war room is one I’ve pushed for over and over. Migrations can be massive changes. Making as many components of the migration process as smooth as you can is critical. Your success as a migration team rests of the simplicity of the migration.

Because no matter what you do a new solution means many more helpdesk calls.

.doc

Scott Andersen

IASA Fellow

The migration triangle continued

http://docandersen.podbean.com
https://docandersen.wordpress.com
http://scottoandersen.wordpress.com
My Amazon author page!!!!
http://lukeoandersen.wordpress.com
http://chuckandersen.wordpress.com
http://NickOandersen.wordpress.com
http://content.iasahome.org/blog
http://www.safegov.org

Yesterday I introduced the three legs of the migration triangle Communication, Planning and people. I am expanding that thought and taking a look at where migrations may fail and relating them back to processes we can leverage to reduce the risk. They are currently in the form of initial rules. I may change that as I go along but for now, we are calling them the MTR (migration Triangle Rules)

MTR

Rule 1: Absolutes don’t work. If you migration starts with the concept of attaining an absolute during the migration for peak effectiveness your migration will never be effective.

Rule 2: Separation between the people designing the migration, people who have done these types of migration before and the people actually doing the migration cannot be great. They don’t have to be in the same city but you can’t have a consulting company design your migration, and leave. The migration designer has to be engaged until roughly the middle of the actual migration to sustain the process going forward.

Rule 3: There are no wrong answers. Then there are wrong answers. The brainstorming rule is critical. In solving problems during the migration its important to remember you start with no wrong answers. You end with only one right answer. Getting to the one right answer involves having the capacity to test solutions against the problem to determine what is best before declaring victory and moving on to the next problem.

.doc

Scott Andersen

IASA Fellow

Introducing the Migration Triangle.

http://docandersen.podbean.com
https://docandersen.wordpress.com
http://scottoandersen.wordpress.com
My Amazon author page!!!!
http://lukeoandersen.wordpress.com
http://chuckandersen.wordpress.com
http://NickOandersen.wordpress.com
http://content.iasahome.org/blog
http://www.safegov.org

The concept of upgrades and migrations has bothered me for a long time. First off because I have been engaged and involved with a number of them.

Problem 1: successful migrations are tied to having smart people on the team. (not me, I just happened to know who the smart people were).

Problem 2: migrations are about change and change is about risk. Risk and change aren’t bad they just have to be managed. Some of the best project managers I’ve worked with grasped this on all levels. Don’t panic, risks happen!

Problem 3: the migration process doesn’t always allow for the existence of the previous version. For example several times during upgrades you find that what worked before now no longer works. It doesn’t mean what you built before was bad, it just means you have a migration within a migration.

The reality of the three problems is that I haven’t just discovered them. The school of hard knocks, degree in migrations wasn’t just put on my door. In fact I’ve walked into every migration I’ve ever done knowing these rules and sharing them with the team.

They get you every time however. Never all three (if all three bite you really have a bigger problem than planning and communication) at the same time. The first one is solved by picking and engaging the right people. The second one is all about communication and the third is all about planning.

I would like to call that the migration triangle. Communication, Planning and people are the three components of this migration triangle. You can lock any one of the three, but you cannot lock all three. If you say we are doing a project with no communication, then you have to have exceptional people who plan like crazy. So on and so forth, you can lock any one side of the migration triangle.

.doc

Scott Andersen

IASA Fellow

What if your exit on the information superhighway was just a manual typewriter?

http://docandersen.podbean.com
https://docandersen.wordpress.com
http://scottoandersen.wordpress.com
My Amazon author page!!!!
http://lukeoandersen.wordpress.com
http://chuckandersen.wordpress.com
http://NickOandersen.wordpress.com
http://content.iasahome.org/blog
http://www.safegov.org

25 years ago I was a school teacher. I used to get in trouble because I was using an electronic grade book, that let me compile my end of year, end of semester grades very quickly and get them onto the permanent record the morning of the first in-service day. That left me with an entire in-service day and 1/2 of the first day to engage in other projects. I saved a ton of time.

In 1989 I did my taxes on the computer, which saved a ton of time but still had to print them and mail them. In the end that was a time sync. I don’t recall what year the IRS started allowing online submissions but I did that the first year it was available. I saved a ton of time doing my taxes.

For me technology has been a time wasting time saving journey. Things that I start doing at first using technology solutions take more time but eventually they begin to take less and less time. My typing skills (honed in my father’s suggested typing class in High School) continue get faster and faster. I should make my sons take typing. The computer saves me a ton of time.

So what do I do with the time I save? I spend more time with technology. It is an interesting balance between the potential time savings and the reality of the lost time seeking other answers.

The adoption rate of technological solutions to complex and simple problems represents a curve. The more time you save from the overall use of the solution the more likely you are to continue using the solution going forward. I couldn’t imagine doing my taxes on anything other than TurboTax (and then having an account check it). The hours saved by simply organizing taxes over the course of a year makes your life much simpler.

I wonder how easy it would be to do my blog in a non-technical manner? Typing it on a manual typewriter, scanning it into the computer and then posting it.

What if your exit on the information super highway was a manual typewriter?

.doc

Scott Andersen

IASA Fellow

The tyranny of titles.

http://docandersen.podbean.com
https://docandersen.wordpress.com
http://scottoandersen.wordpress.com
My Amazon author page!!!!
http://lukeoandersen.wordpress.com
http://chuckandersen.wordpress.com
http://NickOandersen.wordpress.com
http://content.iasahome.org/blog
http://www.safegov.org

I know some people in various companies I’ve worked with over the years who have a concept of a title and drive towards that title relentlessly. They are, however bound to that title and nothing else.

To them the broad concept of title is more important than virtually anything. If someone else gets the title and they don’t have it they question that person (he can’t be an blah de blah he doesn’t even know how to tie his shoes).

I also know people who are bound by job. They want a specific job (title not required). It’s the same as those bound by title as well, and they also react negatively when someone else gets the job they wanted.

The reality of the world around us is there is plenty of work and precious little glory (the right job, the right title). We spend so much time consumed with that reality that in the end we forget we have things that have to be done.

I am not sure if the tyranny of titles is an anti-pattern yet for communication. It certainly isn’t the best possible outcome for a team but I am not sure it directly impacts communication.

In reality it really isn’t one to worry about. Other than the consumption of the people who are driven by it. You do have to feel sorry for them at that point.

.doc

Scott Andersen

IASA Fellow

Did we buy our last DVD?

http://docandersen.podbean.com
https://docandersen.wordpress.com
http://scottoandersen.wordpress.com
My Amazon author page!!!!
http://lukeoandersen.wordpress.com
http://chuckandersen.wordpress.com
http://NickOandersen.wordpress.com
http://content.iasahome.org/blog
http://www.safegov.org

84 down and 39 up with FIOS today. 9ms latency. I remember connecting to work with a 14.4 modem and praying no one would pick up the phone in the house.

There was a time when we had one computer in the house that dialed up to the Internet. Once we got ISDN in 1999 we had more computers connected but nobody was bragging about the speed. For awhile I had a router that handled DSL and Cable connections at the same time, so we had a little more bandwidth. The best rating I ever got in Indiana on the cable modem was 25 up and 15 down. Once everyone in the neighborhood got cable modems those numbers went down radically.

We moved to Maryland but continued with a cable modem. We were Comcast in Indiana at the end and Comcast in Maryland. We were in a little cleaner space in Maryland and ended up getting 51 up and 22 down. We were paying for more but never did better than that.

Roughly 14 months ago we switched to FIOS. It has been an amazing difference. I used to make sure I downloaded large files while connected at work, now I simply download large files when I need them. I can pull down images and other files at will. What a difference home bandwidth makes.

It makes everything you own just a little more useful.

There was a time when we had one computer connected. Now we have around ~25 devices connected at various times. We are still limited to 3 Netflix streams (performance degrades at that point) but still that is infinitely more than we could have ever imagined just a few years ago.

I always said the Internet would take off with fast Internet connections at home. The same was true of cloud computing. Now the question is with all this bandwidth will we ever buy a DVD again?

.doc

Scott Andersen

IASA Fellow

How do we move to the cloud

http://docandersen.podbean.com
https://docandersen.wordpress.com
http://scottoandersen.wordpress.com
My Amazon author page!!!!
http://lukeoandersen.wordpress.com
http://chuckandersen.wordpress.com
http://NickOandersen.wordpress.com
http://content.iasahome.org/blog
http://www.safegov.org

The forgotten change.

I woke up the other day thinking about a customer question – how do we move to the cloud. Its an easy question right? One I answer every day. But it is a question that begs many more questions. In fact it demands some questions.

What are you moving?

The first question of course. The reality is anything can be virtualized. Anything can run in the cloud. That’s a fact. There is no difference in the end between a virtual machine and a physical machine. Now that said there are performance goals that virtual machines will struggle to meet mostly in the area of video performance. But that remains a predominantly VDI issue. The what of your migration is really important.

Depending on the answer to this initial question we have a number of other path’s and questions to ask. This is however the critical question to start with.

“We would like to move our email.”

Or collaboration or, well any of the traditional cloud solutions. Those are easy to plan for. Migrations are painful so you have to have a good partner to lean on, but they aren’t rocket science so we can end up in the right place.

“We would like to move some of our business processes.”

Oh boy, now the questions will flood out of you.

(more to come)

 

.doc

Scott Andersen

IASA Fellow