Everyone’s very familiar with server virtualization technology (VMware, Hyper-V, Xen, KVM); creating virtual server instances instead of matching server hardware up with a single operating system. Most medium and many small sized business have carved out a server virtualization strategy and are well down the path of virtualizing the computer room or data center. Virtualizing servers is an established best practice.
What you may not realize is the heated battle between vendors to virtualize the network is in full gear (network virtualization or Software Defined Networks). Combatants include well established companies such as Cisco, VMware and IBM, open source initiatives like OpenStack, university research programs, and industry organizations and standards including OpenFlow (ONF), NfV (ETSI), and OpenDaylight (Linux Foundation).
Consider network virtualization an open battlefield where new entrants and established players see an opportunity to unseat traditional network vendors, most notably Cisco. During August 2013’s VMworld Expo, VMware pre-announced their NSX network hypervisor. HP and Juniper co-announced plans to integrate VMware NSX with their respective SDN controllers and Layer 2 gateways.
Keeping up on all that’s happening in network virtualization can be challenging for network engineers and IT managers in small-to-medium sized businesses. What’s occurring is more than just an incremental move to virtualize elements within of the network. SDN and virtualization are fundamentally redefining how we design and think about data networks, emphasizing software functionality over traditional network hardware, dynamic network creation/reconfiguration through OpenFlow and network controllers, collapsing network functions into multi-purpose network devices, and performing network administration tasks through APIs and scripting languages such as Python and Django framework. These represent substantial, if not fundamental, changes in how we design and manage networks today, and the network engineering skills necessary.
What should IT and network engineering organizations do to prepare for network virtualization?
Short URL to this post: http://goo.gl/f6eGkm
This post was written as part of the IBM for Midsize Business program, which provides midsize businesses with the tools, expertise and solutions they need to become engines of a smarter planet. I’ve been compensated to contribute to this program, but the opinions expressed in this post are my own and don’t necessarily represent IBM’s positions, strategies or opinions.
Amazon Web Services, Cisco, Cisco Nexus, Django, HP, Hyper-V, IBM, Juniper, Kemp, KVM, load balancer, network hypervisor, network virtualization, NfV, open source, OpenDaylight, OpenFlow, OpenStack, Python, SDN, SDN controller, Software Defined Networks, virtualization strategy, VMware, VMware NSX, web services, Xen
Welcome again to the Alan and Mitchell Podcast. We're still working on a new podcast name so hang in there until we have something better.
During the podcast, Alan and I talk about:
1. New security features in Windows Azure
2. The first zero day for Windows 7
3. 2nd iPhone worm
5. Why is security so hard
6. Fortinet's IPO
7. Cloud Computing becoming part of the pop culture
Thanks for joining us and enjoy the podcast!
One thing that happens to everyone at one time or another is when you become so engrossed inyour own world view, you start to believe everyone else thinks the way youdo, or if they don’t, your spin will fool them. Doesn’t matter whether you’rebig like Cisco and Microsoft, or the latest startup on the block with a newmouse trap. You hear phrases like "he believes too much of his ownpress" (I’m sure that’s been said about me more than once, lol) or"they’ve been breathing their own exhaust too long." I blogged aboutwhat could be one such case of this, Microsoft’s self makeover to be perceived as "open sourcefriendly". Another example is Microsoft claiming it supports Linux in Hyper-V, but only if it’s Novell’s SUSE Linux.
I’m a big believer in ideas like enrollment, passion and engagement, and toachieve these you have to believe in what you, your product and your company aredoing. Doesn’t matter if you are the press spokesperson or the person answeringthe customer service phone — everyone else can pretty easily tell if you areenrolled in what you are doing, or it’s more a matter of your going through themotions.
But that same passion and engagement can also create a blindness, especiallyin entrepreneurial environment where passion, ideas and commitment runs high.It’s easy to build a wall around yourself or your company, focusing just onwhat’s happening inside your product, the product development efforts, or eventhe geographical market area where you are physically located. My very goodfriend AlanShimel used to frequently tell me "you need to get out of Boulder moreoften", not because Boulder isn’t a good town (check out Brad Feld’s blogpost about Twenty-Five Square Miles Surround By Reality), but to reallyunderstand the industry, competition, customers and the market.
I recently blogged about (tangentially as it relates to partnering) the bubbleeffect that can happen in a startup company. It is very easy to become soengrossed in what you are doing, crafting your marketing messages, building theproduct, training the sales force in the ways you want them to sell, that youforget there are other people out there. Companies may claim to already do whatyou do, cover the same supposed differentiators, or have already beat you to thepunch but you just don’t know it yet. I call this inward looking focus, "staringat your own bellybuttons."
There are many things I’ve found helpful to me to try and avoid this. In aleadership role you probably have more opportunity to take advantage of thesebut I believe in any of our roles you can find a way, or even ask to participate inthese kinds of activities. Here are a few ideas.
1. Never turn down an opportunity to talk to a customer. Doesn’t matter ifthey are a sales prospect, an unhappy customer who wants to scream at you, orone that’s nicely tucked in and happy. If you have a chance to talk with or meetwith a customer, always, always do it.
2. Support your company’s trade shows and marketing events. I learn more about the industry atmany of the trade shows I attend than I probably do by reading about companiesand the industry online. Even if you aren’t one of the marketing dudes or dudets who normally cover these events, ask to go and help out. Stop by everyone’s booth, introduce yourself, listen totheir pitch, ask questions and learn. It’s so invaluable.
3. Be well read. Read everything you can get your hands on. I get between 30and 60 Google alerts each day. That’s in addition to all the email and blogreading I do. I don’t read them all, just the ones that really catch myinterest, are newsworthy, are something new, or are on a topic I follow. Readblogs, news sites and portals.
4. Inject what you’ve learned. Share it in meetings, on calls, in productdiscussions, in planning discussions, with customers, etc. Bring thatinformation to everyone. Forward relevant info (but don’t spam) to others inyour company. Add your comments/insights up front so they know whether thearticle is worth the read or the value is in your insights.
5. Talk to every company, not just the ones you like. Go talk to yourcompetitors. You might find out they could actually be your partner. Or, theymay still be your competitor. But go meet them. As Alan also told me many times,"stay close to your friends, and even closer to your enemies."
These ideas are pretty basic and simple, and while they might not shake upthe world, they could redefine how you view your own business.
Hey, it’s time for another StillSecure After All These Years Podcast! This week, a special treat. I invited fellow Network World blogger Brad Reese to join Alan and me as our special podcast interview guest.
Brad blogs for Network World Cisco Subnet where he covers all happenings in Ciscoland. In addition to blogger Brad also has his own company which repairs Cisco equipment.
Out interview with Brad covers a range of topics but one area I think you’ll particularly enjoy is Brad’s views on Cisco’s new CTO. Brad’s received some flack for his controversial criticisms of former Motorola CTO , now Cisco CTO, Padmasree Warrior. But neither Alan or I let Brad off easy – he has to back up his views about Warrior and I think Brad fairs pretty well. Both Alan and I respect people with strong opinions and the depth to back it up so you’ll find this discussion very engaging.
Enjoy the podcast and feel free to drop us any suggestions or questions at firstname.lastname@example.org.
Coming Tomorrow, January 1 – Make VoIP calls on the iPod Touch.
Three French developers have created a hack for the Apple iPod Touch thatwill allow it to be used as a VoIP phone. I have to assume they’ve added the SIPstack to the Touch and are using the WiFi connection on the iPod Touch to makeVoIP calls.
Will Apple squash this French iPod VoIP Revolution and make bricksout of iPod Touch devices? Possible but since there’s not a required ATT serviceagreement for the Touch like the iPhone there’s not the same networkrestrictions.
Interesting idea, I can see the tagline now. Let your Apple iPod Touch reach out and touch someone! Ha, those craaaazy frenshhh peep-pulz!
I put a blog post today on my Network World blog about Microsoft turning intoanother Cisco, that is, Microsoft has ramped the acquisition engine and is nowbuying its way into product categories to keep up with more innovativecompanies. I’m thinking specifically of Google.
A commenter to that post reminded me that even DOS was an acquisition, whichis true. I think the difference I’m pointing out here is that Microsoft is nowplaying a game of reactive cat-and-mouse. Google makes a move, Microsoftacquires a technology or company so Google doesn’t gain the upper hand.
Will Microsoft become so bloated they’ll drop under their own weight? Havethey already reached that point? Well, Microsoft still as dominant as they’veever been so it’s not time start call out "the sky is falling" by anystretch.
Cisco announced the I-Prize competition, funding of up to $10 million for the team submitting the winning idea that could make Cisco $1 billion in a five to seven year period. The announcement was made at the opening of their Globalization Center East in Bangalore, India.
It’s a new approach for Cisco who has proven time and again they are an acquisition juggernaut. If this generates good enough ideas it could represent yet another avenue for acquiring innovation created outside Cisco. Interesting enough is that it will also expose Cisco to ideas that don’t win the prize, offering multiple options for gaining external innovation.
$10 million though is not much when it comes to creating products, especially if they are going to be subsumed back into a large company like Cisco at what could prove to be too early a stage. Cisco risks bringing the new technologies in house too soon rather than let them incubate and prove out more fully as the startups they acquire typically do.
Either way, it’s a novel approach for Cisco and will be interesting to see what shakes out of the competition.
If you were doubting whether WiMAX/802.16 would materialize as a broadband wireless access service, Cisco put their $330m where their acquisition machine is by gobbling up WiMAX antenna and base station maker Navini Networks. The prevailing argument (including from Cisco) is that WiMAX eats into existing WiFi’s products’ revenues, but it’s inevitable that carriers offer a wireless broadband network with much broader coverage over longer ranges. Carriers aren’t going to leave broadband’s wireless "last mile" revenues just to cell phone networks. Cable modems and DSL need a healthy wireless competitor and WiMAX is best equipped to bring it on.
Predictions are that WiMAX will make most of its inroads in developing countries where there’s much less wired and fiber infrastructure. Maybe so but there’s no reason WiMAX can’t make big inroads in North America where we all depend on intermittent low-WiFi 802.11 hot spots and cell phone networks for Internet access on PDA phones and tethered PDAs and laptops.
I tether my PDA phone and laptop all the time but would much prefer to ditch the "last 2 feet" of wire between my laptop and PDA phone, for a WiMAX standard in my laptop that could ride on broadband wireless offered by multiple carriers. Bring it on, Cisco. We want WiMAX.
Are we ready for the Cisco/VMware virtual data center? Get ready because that’s the message about the data center’s future from Cisco and VMworld 2007. Essentially, grid computing, storage and networking with VMware as the data center OS and Cisco’s Vframe managing the data center network cloud. Cisco describes this all as the "data center fabric". It may sound dreamy right now but if Cisco is really able to pull off this software strategy, it would literally change the face of the data center.
Stuck right in the middle of this strategy is the shifting sand of Microsoft’s virtualization licensing restrictions, which has had VMware up in arms for some time. Microsoft’s ability to control when it’s OS can be virtualized may allow them to charge for licenses but restricting the technology that can be used raises anti-competitive vibes. Linux obviously does not have these same Microsoft complications but it would take a tectonic plate shift to completely displace Microsoft from the data center. Not in our lifetime anyway. I still see the major sticking point as to whether Cisco can pull of this software coup, it truly would require Cisco to become a very strong software company.
A huge complication no one is talking about is how the "data center fabric" intersects multiple disciplines across IT. Network design, provision and management mashed up with data center server and applications management, software development and system administration sounds like the Pittsburgh Steelers Kordell "Slash" Stewart equivalent of the IT career path.
This organization upheaval, and IT skills and change of job descriptions will be an even bigger challenge than any of the technology issues Cisco is likely to face. Microsoft has made much stronger inroads into networking than Cisco has into IT systems. This game isn’t over so I wouldn’t assume creating a happy ending for Cisco will be an easy one.
It’s 10pm… and do you know if your VoIP network is safe? Just like any other kind of the technology, it’s all a matter of who’s hands tools are in as to whether they are used for good or bad. That’s no different for the open source VoIP penetration testing tool VoIP Hopper, as this Wired story describes how easy it is to hack into many VoIP networks.
VoIP hopper is roughly the equivalent of an early Nessus scanner for the VoIP world. Using VoIP Hopper you can simulate the interactions between a VoIP handset and PIBXs. In the Wired article even basic security such as MAC address filtering wasn’t being used on VoIP networks they broke into. Unfortunately, again, VoIP is like so many other new technologies that are widely deployed but security is an afterthought. If someone told you we were going allow some new strange mobile device onto the network, we’d at least take a second look wouldn’t we? Many VoIP networks still consider the wired network inside the firewall a "trusted" network but the opposite is really true.
If you are using VoIP in your network, you should at least be firewalling VoIP traffic through a firewall setup sepcifically for SIP and VoIP call handling, preventing access to other data and servers. VLANs and MAC filtering can help but aren’t a cureall. VoIP is another reason to get your network into the 802.1X era so handsets have to authenticate. Bottom line, don’t take VoIP security lightly. Voice is likely your most mission critical buisness application. Bring down voice and email, and many businesses are severely crippled.