You are currently browsing the category archive for the ‘Internet’ category.
Comcast announced bandwidth caps for broadband Internet customers today. After October 1 customers will be capped at 250 GB of traffic per month.
What does 250 GB of traffic mean to the typical Comcast customer? I’m sure that most of them have no idea what it means. Let’s try to put it into perspective.
250 GB of traffic means about 125 standard-definition movies per month (but who is watching SD on their pretty HD plasma screens these days?)
250 GB of traffic means about 25 HD movies per month or around 50 hours of HDTV per month (1.6 hours of HDTV per day).
250 GB of traffic means running a T1 connection (1.5Mbps) around full throttle for an entire month.
According to this interesting article if U.S. broadband Internet speeds continue to change at the same rate as this past year we won’t catch up to Japan’s broadband speeds for another 100 years. It kind of puts our plight into perspective.
After being exposed to the powerful rays of cloud computing marketecture for three days at HostingCon in Chicago I thought I would be granted a reprieve back home in Minnesota. But people are still asking me about this cloud computing phenomena. How will it change our hosting industry? Are SMB’s gravitating towards this service model? Can these cloud computing grids really scale?
I love the thought of true utility-based, grid-architected, infinitely distributable computing services. But what you are seeing marketed today ain’t all that. It’s not even close.
An experienced industry expert told me that cloud computing encompasses any web-based service that can be hosted on the Internet. Well if that’s the case then it’s been cloudy around here since at least 1994.
All of the current cloud computing hosting services are based around the same theme. Take a computer server and virtualize it. Now add another box, some network storage, a pretty web control panel, tie resource metrics to a billing system and voila… cloud computing. Okay, maybe it’s a little harder than that. But I don’t think this is much of an oversimplification.
I recognize that all of this is simply the first step towards the ultimate goal: an architecture that is truly unrelated to the services it provides. I’m willing to wait for that. But in the meantime I’ll keep my head in the clouds.
My good friend Graeme Thickins landed on the front page of the Star Tribune business section this morning promoting one of his recent start-up gigs, DoApps. The company released a popular application which turns an iPhone into a flashlight, strobe light, or night light. I’m envious of the article placement. Well done Graeme!
Amazon’s S3 storage service was down for almost 6 hours this past weekend. Just imagine you were running a SaaS company which used S3 for all its storage. Ouch! On the other hand is this level of uptime any worse than what you would expect from any mid or low-tier storage vendor? You get what you pay for.
If you’ve been in IT long enough you probably remember the famous motto stated by Sun CEO Scott McNealy in the 90′s: “The network is the computer”. McNealy’s promise was a decade ahead of its time. I think it is safe to say that today the network is a big part of the computer. But is the network ready to fully supplant our local computing devices? Despite recent proclamations that “everything is a service” I believe that technologies such as cloud computing have a long way to grow before replacing traditional computing architectures.
A number of trends are pushing the development of cloud computing:
– Consumer technology innovation is pushing business technology innovation
– Businesses are adopting collaboration and team-oriented work methodologies
– Cloud computing is driving down CapEX and OpEx for businesses
– Barriers to entry such as connectivity, reliability, and security are decreasing
These trends are great for small businesses and start-up companies. But I believe that cloud computing faces a number of adoption challenges in medium and large enterprises. Here are just a few:
No standards. All the big guys (Google, Microsoft, Amazon, etc…) are building out cloud computing platforms today. The problem is that all of these application publishing platforms are different. I can easily move my Windows applications from a local Dell server to a local HP server. I can’t move my application from Google’s cloud to Amazon’s cloud. I can’t see enterprises betting the farm on a single cloud computing platform.
Legacy code. Large enterprises are running legacy code written decades ago. Cloud computing does not offer the cost savings to justify a rewrite of these legacy applications yet.
Performance. A cloud storage technology like Amazon’s S3 will not come close to matching the performance of local storage resources like a NetApp SAN in the foreseeable future. The use of cloud storage is limited to applications with low performance requirements.
Security. Cloud computing and storage vendors will have a tough time convincing large enterprises (and their auditors) that data is securely managed in their cloud infrastructure. I work with businesses every week that dismiss shared computing architectures in favor of dedicated computing resources. Until businesses are comfortable “crossing the streams” we won’t see large adoption of cloud computing by enterprises.
I’ve been hearing buzz the past few days about a huge flaw discovered in the Internet DNS (Domain Name Services) protocol. Apparently this flaw would allow attackers to perform cache poisoning on DNS servers. All world-wide DNS servers contain this flaw. I asked three different technology people today if they had heard about this DNS flaw yet. None of them had heard about it so obviously the message needs to get out.
DNS cache poisoning is serious stuff. If this flaw was widely exploited it would cause untold worldwide financial damage. Imagine being able to route traffic destined for a website to any place you want. You could siphon off user logins to savings and mutual fund accounts. You could disrupt online retail and B2B transactions. You could direct Obama website visitors to the McCain website. You get the picture.
So far security researchers are keeping quite about the exact DNS cache poisoning attack vector. All DNS servers world-wide will need to be patched. Wow.
Why would Comcast actively block legitimate traffic on their Internet networks? (otherwise known as traffic shaping) I think the reason is pretty simple. Let’s walk through my thought process together.
What is the primary driver of Comcast’s revenue? Answer: video.
What is the primary driver of increasing bandwidth utilization on the Internet? Answer: video.
What is an Internet protocol like bittorrent mainly used for? Answer: video distribution.
Comcast is simply trying to disrupt video distribution on the Internet before the Internet disrupts Comcast’s primary video distribution model. It’s like telcos trying to add latency to 3rd-party VOIP packets in order to diminish the quality of voice calls. These providers will fight back if you threaten the golden goose.
The Star Tribune loves municipal wi-fi. Okay we get the message. You guys on the editorial board can keep pushing wi-fi in Minneapolis all you want. I have no doubt that the city is going to accept the network because they have to save face. The CIO ain’t going to risk her job over this. At PUSH this week not one, not two, but three different people told me they dropped wi-fi because it was way too unreliable. I hear this same story at least once a week. I know this service either works for you or it doesn’t. I’m not dumping on the USI guys because I believe they are trying hard to make this model work. The problem is that I think this technology is going to be short-lived. I wonder if the city has any provisions for getting out of this contract within the next 5 years?
I wrote an article for the Minneapolis Star Tribune titled Minnesota: An Internet Backwater Soon? I can’t lay claim to the article title but I think it’s kind of catchy. I’ve received positive responses from friends and colleagues. The article was sitting in the editors hands for months so I was happy to finally see it printed. Minnesota like many other regions of our country is facing bandwidth challenges. The major carriers are not stepping up to the plate. We need change.
