Category: Tech

What is the easiest way to account for space/power consumption of a network element?

Am making a spreadsheet comparing different products and looking at longer term costs, maintenance, power, cooling, etc.  I felt that rather than scrubbing the DOE sites and trying to get power costs by state I would just use the national average, but then fell flat on that because I found negotiated rates could be much less than published tariff rates.

Then I stumbled upon what may be an easier solution to my quandary and one inline with what I see a lot of enterprises doing – call a hosting company.  I haven’t talked to too many enterprise customers that are not at least considering if not seriously considering using a hosting environment, or event a full-blown cloud deployment for some portion of their enterprise data center workloads.  Why? – the main reason I keep hearing is that most enterprise customers cannot build big enough to achieve the same economy of scale as a Google, Microsoft, Facebook, etc.  So they may as well lease space from a provider who can achieve a higher density, lower PUE, better delta-T, and handle the compliance tasks like SAS 70 Type-II (Switch, Equinix, Corelink, etc) and not to mention the IT assets put within the data center grow at a power/performance curve that usually breaks the facility they are housed within in 5-7 years, so who wants that on their books – better to let the provider manage/operate it.

In asking around I got to an average number of ~$155 per month per kilowatt consumed when in a hosted environment (ping, power, pipe).  Does this seem inline to you or too high/low based on what you are seeing?

With this data you can then extrapolate Watts/10Gb port across several systems and you get variability from $92/year per 10GbE port up to $372/year per 10GbE port assuming $155/month per kilowatt.  (I am eliminating my own companies products from this so as to avoid being a blatant advertorial…) Annualized hosting/power cost comes to $9,400 to $25,800.

I will be the first to admit there are HUGE differences in features, programmability, buffering, network segmentation, encapsulation methods, and Quality of Service Granularity between many of these platforms.  Those that performed the best were usally more ‘switch like’ with smaller buffers, less features, and fixed function ASICs for the data path.  Those at the top end of the spectrum were almost always products like Juniper’s T640/T1600 and Cisco’s CRS – extremely high function core routers with huge performance, buffers, shapers, policers, and probably most importantly a software upgradeable packet processing engine that allows incremental feature additions that execute in the data plane.

It’s clearly not an apples-to-apples and don’t want it to come across that way, my real question is – is using an average of US hosting pricing per kilowatt an effective way to get a model for opex cost/10Gb port or are there other models people would recommend?  Am pretty open to anything right now provided it is accurate and neutrally intentioned.

dg

If you don't like synthetic astroturf, you'd plant real grass, right?

In March 1999 Scott Bradner from Harvard University and Jim McQuaid of NetScout got together and published RFC 2544 – “Benchmarking Methodology.”  In the subsequent eleven years this informational RFC has been used to provide a baseline for testing many networking devices.  It is designed to provide consistency between vendors so an end-customer can make a more informed buying decision and have some idea of the performance and scalability characteristics of the products they are considering.

For many years this RFC was applied by testing companies to provide comparisons and contrasts between different networking vendors.  Recently though, a company who usually takes an ‘elder statesman’ role in the networking industry and takes pride in its public brand image wrote that this was ‘synthetic testing’ and was not in any way indicative of ‘real world’ performance results customers were likely to see.  This was published on their blogs, and then on comments made on NetworkWorld’s web site by their employees renouncing the testing and trying to invalidate the good work of David Newman.

I have a simple question…

“In the last eleven years why didn’t you write a better and more ‘real world’ benchmarking methodology if the one you blast as synthetic is really that deficient?”

I mean, let’s be serious, you are a huge company, and have the resources.  You have lots of people who go to the IETF meetings and try to steer standards.  You have lots of customers and have no problem telling us that, so it can’t be a lack of revenue.   Why not just help us all by writing a better test plan rather than proverbially taking your ball and going home?

As I close this little diatribe let me remind everyone of two fun little stories…

In 2006 Kanye West was up for ‘Best Video Award’ at the European MTV Music Awards.  He won in a smaller category, ‘Best Hip Hop Artist’ but failed to win the prestigious ‘Best Video Award’ losing to a  smaller production.  He stormed the stage and “lashed out in a tirade filled with expletives,” West said he should have won the prize for his video “Touch the Sky,” because it “cost a million dollars, Pam Anderson was in it, and I was jumping across canyons.”

Apparently to the judges it didn’t matter how much Kanye spent, or that he looked cool flying across canyons, they judged on value.

By contrast at the 2009 Academy Awards ‘Slumdog Millionaire’ won Best Picture, Best Direction, and six other Oscars.  As Danny Boyle and then Christian Colson took the stage to thank their teams and supporters their competitors stood up and cheered for their victory.  You never saw Ron Howard, Gus van Sant, or Sydney Pollack trash-talking the Academy for how they voted.

These guys are smart enough to know two things – One, you are measured by how you well you lose as much as by how you win.  Two, if you bad mouth the Academy how will they treat you next year?

Do you want your primary networking vendor to be more of a Kanye West or more of a Ron Howard?

Imagine a car that was fixed speed and couldn't adapt from highway to city to parking lot. Just sayin'...

I am still a very proud alumnus of cisco Systems, but am also not bashful about areas I think the networking behemoth can improve.  My main recommendation would be to get the business units working together to consistently solve customer problems – be a big company, but act like one company, not 20 or 50 or however many initiatives, boards, councils, or work streams there are.  As a former commissioned combat arms officer I will state that some things are better run in a command and control environment if you want consistency and necessary if your customers want a consistent experience.

The recent ‘data center’ announcement of 10GBASE-T products really served to illustrate this better than I could ever explain.

According to Cisco’s Frequently Asked Questions about their 10GBASE-T products for the Catalyst 6500 and 4500 they state the following:

Q. Is the 10GBASE-T line card on the Catalyst 4900M compatible with Gigabit Ethernet?
A. The 10GBASE-T line-card module on the Cisco Catalyst 4900M supports Gigabit Ethernet or 10 Gigabit Ethernet mode for each port group. The eight ports are divided into four port groups, and each port group can be configured to operate in either Gigabit Ethernet or 10 Gigabit Ethernet mode. All ports within the same port group must have the same mode. This allows customers an easy migration path from Gigabit Ethernet to 10 Gigabit Ethernet network connectivity.

Q. Can the 10GBASE-T line card on the Cisco Catalyst 6500 Series be connected to Gigabit Ethernet network adapters using auto-negotiation?
A. No, the 10GBASE-T line-card module will not support Gigabit Ethernet. It will support 10 Gigabit Ethernet network adapters only.

Can someone explain to me how the board, or council or whatever new-age org model is in charge decided that customers want a 10GBASE-T port that would support your existing cable plant but not interconnect your existing GbE attached servers on one switch (the Catalyst 6500 – hard coded into the PHY so it is not a field or software upgrade) and wanted a completely different behavior on the other (Catalyst 4900 that did the rather obvious feature of speed autonegotiation)?

Quick car analogy, since some people who have issues with ‘synthetic tests’ (apparently auto-negotiation tests are synthetic now too….)

Why build a network equivalent of a Bentley Continental GT that goes either 12mph or 120mph yet requires a mechanic to switch between the two speeds, and then have the audacity to claim that this offers an easy migration path between the city street and the autobahn? (analogy credit to Ed, you know who you are!)

This makes no sense to me.

dg

Everything seems to come full circle in IT...

Cautious optimism is a term I have been having many discussions lately with friends and analysts about – whether we are seeing true economic recovery or a bit of a ‘W’ and whether to make serious investments in planned growth or not.  Candidly, in IT we have compressed capital spending for a while, so it could just be a bit of elasticity – although one major thing strikes me as different.

In the current world order many of the IT investments seem to be directly proportional to short-mid term ROI, sure everyone wants to build for 5-10 years, but they also want to see real business results, right now.

Mostly this means that new project types are getting priority and IT is finding creative and innovative ways of delivering near-term business value without, hopefully, taking their eye off the architectural ball.  Ideally we can do both- deliver short-term value creation, while building towards a longer-term vision that enables IT to reinvent itself and infrastructure to transcend generational shifts. Sadly this is not always the case, some companies and people seem to want to either over-rotate on short-term. Sadder, others refuse to admit the world is changing.  Even worse are those who keep their head in the sand and cannot move at all.  Denying change happens is dooming almost any business to failure, embracing a fickle trend too quickly can be just as painful, and relying on past formulas from previous successes doesn’t always work.

You may wonder where I am going with this.  Over the past thirteen years I have seen a lot of things change and come full circle- Cut-Through Switching, Lossless L2 Networks, Ring Topologies, Hosting/Cloud/Insource/Outsource.  Universal truth – things change and open and experienced minds that can capture this change tend to prevail.

Architectures have to change with the trend, the old way of doing things is not always the best- althought there are always viable lessons to be learned and due respect should be paid to past success.

Looking at networking, especially in the data center there are a lot of architectural changes in play.  Obviously, the changes being driven to effect convergence between Ethernet and FibreChannel is a big one.  The other is the collapsing of layers and efforts to simplify the topologies while increasing the scale of operations – I think in my next post or two I will have to explore these some more, what are your thoughts on other architectural changes in the data center network?

dg