Category Archives: data center network

A green networking idea for cloud computing and data center.


Vern Burke, SwiftWater Telecom
Biddeford, Maine

After commenting on the piece questioning whether cloud computing really saves energy, I came up with an idea that might simplify the whole energy cost of the network issue. Yup, it’s another metric! 🙂

The issue here is, how do you account for the energy cost of transferring data across the Internet to a cloud computing provider? After all, consider the massive amount of networking devices involved in the Internet and that Internet routing protocols are geared towards bandwidth efficiency rather than energy efficiency.

So, why not create an energy based routing protocol? For the sake of discussion, I’ll call this OGPF (Open Greenest Path First). Unlike OSPF (Open Shortest Path First) which looks for the shortest way from point A to point B using weighted “length” metrics, OGPF could use an energy metric for every device in the path, possibly including how the device is powered.

Now you have a network that’s automatically as green as possible and you have a way to at least get a measurable idea of what’s going on across a network wide link.

Advertisements

The net neutrality red herring: Anti net neutrality, Comcast and Level 3


Vern Burke, SwiftWater Telecom
Biddeford, Maine

I’ve been watching the debate over the Comcast vs Level 3 peering dispute. I’m not a big fan of Comcast by any means, but any attempt to paint this as a net neutrality fight is ridiculous.

The net neutrality issue is really very simple. Large ISPs thought it was a good idea to charge content providers for priority paths through their network to reach their subscribers. Pay the “toll” and your traffic speeds right along smoothly and your users love you. Don’t pay the “toll” and you take your chances with a cow path instead of a nice smooth 10 lane superhighway. This kind of extortion is bad in so many ways that I can’t begin to count them.

Peering, on the other hand, is simply an agreement between two providers that they will exchange traffic between them directly, rather than going through the public Internet backbone. This can save both providers by removing traffic from their expensive Internet backbone links and it speeds up the traffic flow between both providers dramatically. No charge peering works only if the peering arrangement is of equal benefit to both sides. This means that the traffic flows roughly evenly in both directions.

My understanding of this issue is that Comcast and Level 3 have had a mutual peering arrangement for some time now. Now Level 3 is demanding special peering access deep into Comcast’s network for the sole purpose of speeding their traffic to the end user. This is an arrangement that solely benefits Level 3 but they insist on Comcast paying the entire cost.

It’s quite clear that this is, in no way, a net neutrality issue. If anything, this is an ANTI net neutrality issue. Instead of the ISP wanting the provider to pay for premium access through the ISP’s network, the provider is demanding the the ISP give them free premium access deep into the ISP’s network and attempting to claim that they are somehow entitled to it.

Level 3 needs to man up and come up with an arrangement that mutually benefits everyone involved or pay the fair cost of the special access they want that benefits nobody but them. Level 3 can throw this at the net neutrality wall all they like, but it doesn’t stick.

Wednesday data center tidbits: VMware tap dances, RIPE and Duke make excuses


The funniest thing I’ve read all week is the piece about VMware dissing bare metal desktop hypervisors. The sequence sort of goes like this:

1. We promised a bare metal desktop hypervisor.

2. Wow, this is harder than it looks.

3. Well, the hardware on the average PC wouldn’t be compatible anyways so we don’t REALLY need to do this.

4. If we WERE going to do it, we’d do it better than those pansies over at Citrix anyways because we’re the experts.

When you consider I can load a bare metal server class hypervisor on regular PC hardware (Xen) without a problem and Citrix already has their bare metal client hypervisor, VMware comes off sounding like a petulant child.

Next up is a follow on to the RIPE/Duke BGP routing fiasco. It’s nice to know that it was a Cisco router bug that caused this “experiment” to go out of control but it doesn’t change the fact that NOBODY should have been feeding “experimental” BGP announcements out to the live Internet, period. Gives new meaning to the term irresponsible.

Email or call me or visit the SwiftWater Telecom web site for green data center services today.

Vern

swiftwater telecom rcs cloud computing logo

Monday data center tidbits: Europe blows up the Internet, “e” this


First up today is the piece about a BGP experiment run by RIPE and Duke University going awry. Apparently they decided to try out their experimentally formatted BGP announcements by actually announcing them to the net, resulting in predictable chaos. Now we’ve moved from taking down data centers with risky action for no gain, we’re taking down portions of the Internet as a whole for an experiment that should have been conducted in a lab. Duke and RIPE, Internet bozos of the month.

From the Twitter stream, “One vendor promoting eClouds at #vmworld Do we still need the “e”? It’s like pushing the eInternet or the iWeb”. “eClouds”? This has to be the dumbest thing I’ve ever seen. Supporting bozo award for worst marketing crap ever.

Email or call me or visit the SwiftWater Telecom web site for cloud computing services (no “e”s added).

Vern

Tuesday data center tidbits: #cloudcomputing regrets?


First up is the piece about EMC shutting down it’s ATMOS cloud computing storage service. Vendors have been inadvertently competing with their own market far longer than cloud computing has been around, this isn’t a cloud phenomena.

Next up is a piece about cloud computing and data center secrecy being an IT security risk. Revealing the physical location of a public data center, offering site surveys, and being willing to talk to some general degree about data center facility and security architecture details is all reasonable. There shouldn’t be an expectation of total transparency without an NDA however (how much of your internal business operations do you reveal to the world in general?). There’s an insinuation that, if you don’t give away everything, that you must be hiding something bad.

Finally, there’s another piece about cloud computing security risks. I have no idea where people get this idea that, because data from multiple tenants is stored on the same server, that this automatically means everyone can get into your data, not to mention that you can’t investigate any security breach because tenant’s logs are “co-located”. It appears to me that most of these people are pontificating about something they’ve never worked with and don’t fully understand.

Email or call me or visit the SwiftWater Telecom web site for cloud computing services.

Vern

Monday data center tidbits: MIT babbles about cheap optical Internet, #cloudcomputing “regulation”


First up, MIT develops an Internet router that’s faster, cheaper, and makes no sense whatsoever. I don’t know if someone at MIT is having a brain cramp or it’s the fault of the journalists, but the idea that routers convert optical signals to electrical because they can’t deal with optical signals coming from different directions is stupid. Routers convert optical signals to electrical because nobody has developed an optical processor for routers. CPUs require electrical signals, hence the requirement for conversion, it isn’t any more complicated than that.

I’ve seen quite a few articles such as this one calling for “regulation of cloud computing”. Yup, this is all we need, get the government involved in regulating all cloud computing providers into uniformity so no one will ever have to bother to do their homework and understand the service they’re buying. That’s like regulating all the car manufacturers down to one because it’s too confusing to understand the differences between a Chevy and a Ford. Foolish idea.

Email or call me or visit the SwiftWater Telecom web site for cloud computing services.

Vern

Friday data center tidbits: Amazon bozos #cloudcomputing (again)(and again and again)


Today I’ve been reading more detail into yet another recent data center power failure of the Amazon EC2 cloud computing service. There isn’t much they could do about having a car crash into a utility pole, but the rest of it is purely ridiculous.

1. The transfer switch that failed came with a different default configuration than their regular ones. What kind of a bonehead installs a critical piece of gear like a transfer switch (or any other piece of data center infrastructure equipment) without checking the configuration? Operate like this and you deserve what you get.

2. This is a perfect example why trying to save money with absolutely minimum run time power back up is a dumb idea. Sure, if everything goes perfectly, 15 seconds to get a generator started is fine. Gambling that you aren’t going to need enough run time to deal with a problem is a bad risk.

3. Excusing this by saying only some customers were affected does not inspire confidence. Just how many more of these misconfigured transfer switches are out there waiting to screw up?

This is the first time I’ve ever had anyone double up on data center bozo of the week. I’ll take suggestions from the floor on what would be an appropriate award for this dubious honor.

Email or call me or visit the SwiftWater Telecom web site for green and reliable data center DC power, bozo free!

Vern