Cisco has killed the Flip. Twitter and blogs are replicating this announcement out to the far corners of the Earth. I am cheering as are plenty of other network centric people.
My friend Tom(@networkingnerd) spun up a quick post on his blog about the Flip. He doesn’t come out and say that he thought Cisco and Flip were a great partnership. He also doesn’t feel the need to pile on Cisco and point and laugh like so many of us are doing now. Tom says:
“I’m going to take a slightly different line of reasoning here. I don’t think Cisco failed with the Flip. I don’t consider something to be a failure so long as you learned something from it. Apollo 13 wasn’t a failed moon landing. It was a successful astronaut rescue. We learned how to think on the fly when the pressure was on and bring people home safely when it counted. In a slightly different way, I think Cisco learned a lot about what went wrong with the Flip and dissecting it over the coming months should yield a lot of information about how to avoid things like this in the future.”
While I agree in principle, I think the bigger problem with Cisco and Flip was focus. It’s not that Flip didn’t have revenue potential. I happen to think it is a cool product! If you read Tom’s post that I just referenced, he includes some good possibilities for Flip enhancements to make it even easier to share video with others. I think the problem was more of perception from end users of Cisco. By end users, I am not referring to the Linksys, Umi, Flip crowd. I am referring to enterprise networks and service provider customers.
The signal that myself and others were getting from Cisco was that they wanted to be all things to all people. They wanted to meet the needs of every sector possible. In my opinion, it had everything to do with growing the company year after year, and quarter after quarter. It was because of that perception, that I think people started looking more heavily at alternatives to Cisco.
Of course, Cisco can give you all sorts of numbers behind how many engineers support this line and that line. They’ll tell you that they are able to focus in all the different sectors they are in. New products are being released for each of these lines. The impression Cisco wants to convey is that they can juggle all of these different things and still produce quality products in a timely manner.
Let’s assume those things are true. Let’s assume that Cisco CAN manage all the various business units and that they are all working in concert to achieve the goals John Chambers has set for them. All of that means absolutely nothing if the perception from the average end user out there is different. Perception is reality for so many of us.
As luck would have it, the recent memo from John Chambers to the employees of Cisco has strengthened the viewpoint of many of us that Cisco needs to get back on track with the things they are good at.
For what it’s worth, I still believe the following:
1. Cisco produces some great products.
2. Cisco has a large number of very smart and talented people working for them.
3. Cisco can still innovate even outside of acquisitions.
4. Cisco has a VERY effective marketing machine that can sell almost anything.
5. Cisco wraps their products around a vast ecosystem that makes it hard to do without.
6. Cisco WILL continue to dabble in the consumer market.
Closing Thoughts
I’ll sum it all up regarding the Flip with this: “Just because you can sell something, it doesn’t mean you should.”
I hope that we’ll see some additional changes from Cisco in the coming year. Ditching the Flip is just the start. There’s too much competition out there right now for Cisco to ignore. Other companies are producing products that are technically superior and cost a lot less in several of the sectors that Cisco holds a lead in. As an end user, I want to see Cisco continue to produce good products. Ultimately, I have a responsibility to my employer to ensure the products we use are the best fit for the company. I can’t wait around hoping for change from vendors. I have to deal with the problems I have today and choose vendors that are focused on solving those problems. I’m not saying that I want to get rid of everything Cisco. I’m just saying that many of us are watching a lot closer than we have in the past.
Am I wrong? There’s a good chance I might be. If you disagree, let me know in the comments below.
Let me start off by reminding you that I work for a corporation that is considered an end-user of technology services. I am not a consultant and I don’t work for a vendor.
I do a lot of vendor research these days. Maybe too much. I have a strong desire to know what each major vendor’s solutions are and try to compare them to each other. I can go and talk to each vendor if I want. I know they will tell me all of the reasons I shouldn’t buy their competitor’s gear. They’ll tell me why I should buy their gear. Sometimes their points are valid. Other times they’re just making stuff up. Opinion. FUD(Fear, Uncertainty, and Doubt). It happens quite frequently. If it didn’t, they wouldn’t sell as much product as they do.
“You can’t go with that vendor because they are new.”
“You can’t choose that product because it is proprietary.”
“You won’t get the attention you deserve from that vendor because they are too large.”
“You’ll pay too much if you buy their product.”
“Our product has much better performance than our competition’s.”
“They have a history of poor product support.”
“They’re using old technology.”
“They can’t support XYZ.”
Blah. Blah. Blah. Blah. Blah.
I would imagine if I was given a preview of sales information behind the scenes of every major networking vendor, I would find that they keep detailed documents on their competitor’s products and how to win deals against them. I know this stuff exists for certain vendors out there. I assume it exists for all of them. The small amount of this type of documentation that I have seen was very insightful. I’m not talking about a slide deck from marketing with their special math designed to make capacity and throughput seem a lot bigger than it really is. I’m talking about the stuff where they tell you how to answer certain questions that may come up in the sales process. They give you the talking points to use when countering statements made by other vendors in regards to your particular solution offering.
Every single time I talk to a consultant or a vendor, I have to remember one thing. No matter how friendly we are with each other, they still are in business to make money. Everything I am told by anyone outside of my organization, I have to take with a grain of salt. You could probably make the argument for people within your organization as well. I have to consider their angle. It doesn’t invalidate their particular argument for or against something. It just might be that they are only espousing the positives of their offering and the negative of their opponents. Of course, the really good vendor reps and consultants can steer you where they want you to go and make you think you came to that conclusion on your own. My overall concern must be for the good of my employer.
Everyone has an angle. Everyone has bias one way or another. Even within your peer groups. Not everyone has experience with a bunch of vendors. Vendors don’t always make it a habit to provide the small details around how their solutions work, and when they do, they gloss over the negatives or don’t mention them at all.
So I ask three questions:
Who do you trust?
Do you take all vendor/consultant suggestions with a grain of salt?
What can vendors do to help us make better decisions?
Humor me. I’m getting older feeling more philosophical these days.
Mark Fletcher makes a great point about phone availability. Tom Hollingsworth has made a similar point in regards to soft phones. The physical phone is something even small children are proficient with. However, if it is software based, or powered off, it isn’t of much use in an emergency to a non-technical person.
Meraki isn’t the only wireless vendor to offer SSID’s on a time based schedule, but it is a neat offering. Firewalls have been doing this for several years now. Why make something available outside of business hours if you don’t have to?
Another great post from Jose at Initial Draft. If you work with MPLS VPNs or just want to know more about them, this is a must read. Anyone studying for their CCIE Route/Switch lab should add this blog to your RSS reader. They have been creating some real good content lately in addition to their CCIE interview series!
Brad Reese put together an interesting overview of the new Arista 7124SX switch. Take note of the comments at the bottom as well. Doug Gourlay from Arista responds and adds a bit more information regarding the technical specifications of the switch. This is great reading if you like the nerdier things, and I assume you do.
Posted inLinks|Taggedlinks|Comments Off on Things Worth Reading – 29 March 2011
I apologize in advance for the length of this post!
Introduction
Quick! Make me a list of parts needed for an enterprise wireless rollout covering a couple of large buildings with a requirement for 2.4GHz and 5GHz radios. Pretend the site survey has already been done(You do still do these right? You better!) and the access point count is 400. What items do you have on the list?
There’s going to be some other things, but just assume you already have switches that are PoE capable. Now, make the same list with Aerohive.
1) Access points
2) Management platform
Notice anything missing? Of course. The controller! That’s Aerohive in about 30 seconds worth of reading. There’s more to it. A lot more, but the main thing about Aerohive that makes them different from most other wireless vendors is their controller-less stance. Why take that stance? Hardware is advanced enough today to allow full feature sets on access points and have them cooperate with other access points on the network.
Benefits of Controller-Less Wireless Network
Imagine for a second what the benefits are to a controller-less wireless network. I can think of 4 big ones.
1) Performance – What is the effect of having all of your control plane traffic and most of your data plane traffic tunnel back to a central wireless LAN controller? Latency. What if all of your data plane traffic was dumped right onto the switch straight from the access point? What if the access point was smart enough to take care of all the control plane functions like authentication, roaming, etc?
2) Cost – Controllers are expensive. You also have to buy more controllers as you increase the number of access points on the network. Add on additional controllers for redundancy and watch your costs go up even further.
3) Scalability – Consider along with cost, that you can only expand your wireless network so much until you are limited by the controller hardware itself. As you add more and more access points to local or even remote locations, you also have to be aware of how many access points your controllers can support. If you don’t have to worry about controllers, you can scale out as needed.
4) Survivability – If my controller dies, my access points are going to have problems. Hopefully, there is a redundant controller waiting to take over. If so, you better hope that redundant controller is sitting in the same data center or it sits in a data center nearby with a fat pipe between all the access points and it. Of course, in a controller-less environment, this isn’t even a concern. Some vendors can negate controller loss to a certain extent(Cisco and Aruba), but you still lose some functionality in that situation.
Wireless LAN Controller History and Application
Maybe you are wondering why wireless LAN controllers even exist. Well, there is a very good reason they were created.
At the time controllers were developed, it was problematic to manage a bunch of autonomous access points across a large enterprise. Controllers made it a lot easier to do that. Don’t take my word for it though. Listen to Bob O’Hara talk about it.
Controllers solved a problem with regards to management and access point hardware limitations, but in doing so, created another problem. Now, all of your access point traffic(control and data plane), or most of it, comes back to a central wireless LAN controller. This becomes an issue with 802.11n and the amount of bandwidth each AP can support on the wireless side. That traffic has to be sent back to a controller. It is important to note that many wireless vendors are starting to leave some of that data plane traffic out on the edge and let the access point switch it locally instead of hauling it back to the controller, but they are still leveraging controllers to do the heavy lifting. Aerohive cuts out the controller entirely. You can read about that in greater detail here.
My Experience With Aerohive
Last Thursday I was sitting in the Aerohive headquarters in San Jose listening to one of the most amped up technical presentations/demos that I have ever been a part of. We didn’t quite reach Steve Ballmer levels, but it was close! The room was small, but not too small. The Twitter feed was projected on one of the walls. The audience, of which I was a part of, was engaged. The presenters were even more engaged. It was a good time for everyone. Why do I bring this up? I want you to understand how excited Aerohive is about their product. I don’t even use their products and I was ready to start cold calling customers on their behalf right then and there!
Okay, so I should confess that I DO use their products at home. I have a pair of HiveAP 110’s that I received from Aerohive as product samples. I decided to replace my existing single AP at home with a pair from Aerohive. In reality, I was only supposed to get one, but Marcus Burton from CWNP was sitting next to me and he already has several at home and gave his to me. Thanks Marcus! I’m the only one in my neighborhood with my own “hive”. I haven’t done the official “war drive” through my subdivision with my Metageek and AirMagnet(Fluke) spectrum analyzer hardware to know for 100% certain that I am the only one with my own “hive”, but this is the Internet and I can say whatever I want with no repercussions. 🙂 You can rest easy knowing that if you ever swing by my house, I can give you guest access via a captive portal, or just give you your own private pre-shared key without having to give out my pre-shared key that I use to secure my own devices. Cue the Shaft theme music.
Potential Issues With AeroHive
1) I won’t consider Aerohive until they have some market share. – I don’t know what that market share level is supposed to be, or for that matter, how to even measure market share. Wireless is growing leaps and bounds. Companies that weren’t customers a year ago may have a full wireless deployment today with hundreds of access points. It seems to me that market share is a bit harder to nail down in the wireless space. Where I will give this line of thought some credence is in regards to funding. If you are not well funded or don’t understand how to run a business properly, you run the risk of going out of business even if you are selling plenty of hardware. It takes more than a good idea to build a successful company!
2) Aerohive is using unproven technology. – This is essentially the “I wish I had implemented this first” argument. If you are selling controllers and relatively dumb access points, you don’t want to embrace the controller free ideology until you can bring your own similar solution to the market. Or, perhaps you are going to keep selling controllers as long as people will buy them. At what point does it become “proven” technology? I’m thinking it has a lot to do with market share when people make this argument. For the record, every large vendor out there has a ton of smart people working for them. The problem is weaving through the processes these companies have in place in order to bring new solutions to market. They don’t have the flexibility of the smaller companies, but can occasionally one-up them due to good acquisitions (ie Cisco’s purchase of Cognio birthed Clean Air).
3) Aerohive can’t scale. – You can make this argument against the controller as well. 802.11n is going to jack up required throughput levels on these controllers to amounts that have never been seen. It would make sense to be able to dump off the traffic from the access point right into the edge as opposed to bringing it back to the controller. Where this argument holds some truth is when you consider the large number of secure tunnels each Aerohive access point might have to make in order to communicate with other access points. I’m not sure what the maximum number of tunnels the Aerohive access points can support is. I do know that there are deployments out there in excess of 3000 access points. I work in a fairly large building with wireless access points that are placed for voice and location services. It’s a fairly dense layout and I don’t even come close to 3000 access points. How many environments are going to require numbers like that? Merchant silicon has come a long way in the last 10 years.
4) They have to sacrifice features for performance. – With no controller, each access point will have to be intelligent enough to manage a wide variety of things. I can think of 2 things you have to dp without today in an Aerohive environment. First, there isn’t a robust WIPS(Wireless Intrusion Prevention System) offering. Second, there is no spectrum analysis offering. I suspect the spectrum piece will be dealt with at some point if there is enough demand, and I am quite certain an improved WIPS offering is in the works. As for other features, they have them. Pretty much anything from making an SSID available on a schedule(ie Guest access from 8am-5pm) to QoS is available.
Closing Thoughts
Aerohive is at the front of the next phase of wireless networking. Xirrus is there too. Others will be coming along, but I suspect for most it will be incremental steps. As wireless continues to grow, and it WILL grow, companies like Aerohive will be attractive for quite a few reasons.
1) They cost less overall than controller based solutions and have a streamlined pricing structure.
2) Their feature set is on par with controller based solutions.
3) Products like TeacherView(which is free) help lock in verticals like Education.
4) Small product line reduces confusion when trying to sell a solution to new customers. There is such a thing as having too much choice in hardware.
5) There are no single points of failure or bandwidth bottlenecks.
6) Management can happen on premises with a local(physical or virtual) Hive Manager or off premises(in the cloud) depending on the customer needs.
I don’t live in fantasy land where there are no problems with every new vendor that comes along. Aerohive will have problems come up. It might be in their support model. It might be with buggy code. It might be with a couple of bad batches of access point hardware. There might even be architecture issues that I am just not seeing. For now though, I like what I see and haven’t come across any one thing that tells me it would be a bad idea to deploy Aerohive in an enterprise network. The thing that scares me about companies with cool technology like Aerohive and others like Riverbed is that some gigantic tech company will come along and buy them and run the products into the ground.
One Final Point
It is only a matter of time before the really big wireless players jump on the controller-less bandwagon. Motorola and Aruba have already made steps in that direction. Aerohive has to continue to close big deals and increase market share while they still have a leg up on the larger wireless vendors. If they don’t, then 2 or 3 years from now when the big boys are pushing the same solutions, it will be an arms race and Aerohive will lose. My hope is that they continue to innovate with extra offerings like TeacherView and stay one step ahead of the bigger vendors. In time, perhaps they will have an equal or greater share of the market than some of the existing companies do.
No. If I am, then my wife must have found the check before I did. I have followed Aerohive for about a year now. I am a fan of good technology, and even more so when it goes against conventional wisdom. I also like rooting for the underdog. I was part of Wireless Tech Field Day which took place in mid-March out in San Jose. You can read my standard disclaimer here.
If you haven’t read my first post on HP Networking, you can read it here. I covered the marketing aspect of it. In this second post, I wanted to talk about the technical approach that HP is taking. However, there was so much information that was mentioned prior to the technical networking talk, that I couldn’t cover it all in the first post. Therefore, this post will be more marketing type content. Sorry for those of you who hate marketing, but at least I have no slide deck to torture you with.
Let me give you a rundown of the 4 different speakers we listened to from HP. I probably should have covered this in the first post. I mention these people just to let you know how much information we had to consume within the several hours HP presented to us. If you want to see the presentations I saw in person at HP’s Executive Briefing Center, you can watch the videos below. It’s 3 hours worth of content between the 2 videos!
Video 1 – Overall HP Strategy with Frances Guida – Start to 31:25.
Next Generation Data De-Duplication with Jeff DiCorpo – 32:15 to the end of the video.
Video 2 – Overall HP Networking Strategy with Jay Mellman – Start to 23:55.
In Depth Technical Discussion on HP Networking with Jeff Kabul – 24:20 to the end of the video.
Throughout the presentations from HP, you REALLY get the feeling that they only look at Cisco as their competition. Everything was framed in the context of pulling share away from Cisco, or doing things better than Cisco. In light of that, it was no surprise when Jay Mellman mentioned that all of HP’s 6 main data centers were Cisco free. I think they are really proud of that fact, and maybe they should be. Is there any better way to show your customers, or potential customers, that you are serious about your networking products than to “eat your own dog food” in your production environment?
Then, it got REALLY interesting. Jay alluded to a recent Gartner report entitled “Debunking the Myth of the Single-Vendor Network” in which Gartner states that it is cheaper to have more than one vendor supply your network gear. Jay mentioned that Cisco got people very lazy about correct network design and that by bringing in a second vendor, it forces an organization to do proper network design. I am going to assume that was a reference to some of the proprietary things Cisco has developed like EIGRP and HSRP.
One of the delegates, Tom Hollingsworth(@networkingnerd), asked Jay what the difference was between proper network design and lazy network design. Tom mentioned that ProCurve had historically been edge centric and that perhaps HP felt that switching decisions should be made closer to the edge as opposed to Cisco who puts more emphasis on the core. Jay stated that Cisco does that because they make a lot more money selling core switches than they do edge switches. According to Jay, when it comes to Cisco pushing core switching, quote: “It is as much a business model as it is an architectural model.”
HP believes they have a better approach to architecture than Cisco. Maybe they feel that way when compared to the other networking vendors, but again, I get the feeling they are only interested in being better than Cisco. They also believe people are going to do more evaluation than they have in the past.
HP realizes they aren’t going to hit a bunch of home runs and get forklift upgrades from Cisco to HP. They are just looking to get a foot in the door. Maybe they will win a few deals outright, but for the most part, they will have to squeeze their way into Cisco dominated networks piece by piece. BMW was a good example for them. What started out as a small wireless project in a few dealerships blew up into HP getting a piece of the BMW enterprise infrastructure. HP isn’t the only vendor to work the “foot in the door” angle. I’ve talked to several networking vendors in the past year and they are all trying this approach. Get a box or two in the datacenter or on the edge and slowly grow their presence over time. To me, that’s the best strategy. Let an organization get comfortable with you. Then, when there’s a problem and a vendor like Cisco cannot solve it, you get to ride in on the white horse and save the day with your product that CAN solve the problem.
With all of this talk of HP believing they did things better than Cisco, an opportunity to ask HP about voice, or unified communications came up and I took it. I asked Jay if HP was going to do anything in the realm of voice. Granted, they have an existing product from 3Com entitled VCX, but in light of HP’s increasing relationship with Microsoft around unified communications, I didn’t have a good feel for what HP was going to do. The voice/UC offering from Cisco is pretty solid from a stability and feature standpoint, so it would be harder for HP to chip away at that sector than it would be in the realm of switching.
HP has decided they don’t want to be in the voice business long term. Jay indicated that with unified communications(ie voice), it is, and I quote: “bifurcating into applications and infrastructure”. Kudos to Jay for using an obscure word like “bifurcating“. To be quite honest, I had to look it up. 🙂 It means “the splitting of a main body into two parts”. HP has taken the approach that voice is nothing more than an application. They want to focus on the infrastructure that provides transport for that voice traffic, but they don’t want to be involved in developing the platforms that manage/create the voice traffic. Their goal is to identify areas like voice that they consider applications and work with third parties. While I tend to agree that it makes more sense to focus on the infrastructure from an HP networking perspective, it seems to me that HP is one of those companies that could actually put out a voice solution that would work. They have all of the pieces to make it happen. Networking, server hardware, applications expertise, etc. Perhaps to do that, it would take several years of development on their part and they obviously want to remained focused on other things.
I have covered everything(minus the storage de-duplication talk) up to the technical discussion from HP. In the next post, I will jump into the nerdier things. There was so much meaty information from the discussions leading up to the technical presentation that I thought I would re-hash the points that I thought were the most interesting. The more time I spend in the industry, the more interested I get in the non-technical things when it comes to the different vendors out there. That’s not to say that I don’t like the very technical things, because I do. I just think that if you are going to devote a substantial amount of time to learning a vendor’s technology(and we all do), you need to make sure that technology is going to be around for more than a year or two. Understanding where the focus of company XYZ is will go a long way in determining what you need to focus on and what you need to let go the way of the dinosaur.
So……next post on HP will be more technically focused and this time I mean it. 🙂
*****Disclaimer: As a delegate for Tech Field Day 5, my flight, food, lodging and transportation expenses were paid for in part by HP. I am under no obligation to write anything regarding HP either good or bad. Anything I choose to write are my opinions, and mine alone. **********
Posted inhp, vendors|Taggedvendors|Comments Off on HP Networking – Part 2(More vision…)
As part of Tech Field Day 5, I got a chance to sit in on multiple briefings from HP. I was very interested to hear about their particular product set and how it fits within the data center. The following are my thoughts on HP’s networking solution.
According to HP, one of the biggest problems facing their customers is that of “IT sprawl”. As a result of this sprawl, silos are created. The servers end up in a server group. Storage ends up in a storage group. The same goes for the network, database, security, and so forth. Silos, in the opinion of HP are a bad thing. They cause you to lose sight of the bigger picture.
I don’t know that I agree with that. Silos in and of themselves are not a bad thing. It takes a fairly high degree of technical ability to oversee just one of those previously mentioned areas in a decent sized enterprise network. I fail to see how you could have anything but silos. I know there are people out in the industry who think architects should not have a specialty and should be able to design anything at a high level. I call those people crazy. As you go further down the chain into engineering, support, and implementation/deployment, the level of technical abilities in a specific area becomes really important. It isn’t realistic to have people functioning within multiple silos unless the level of technical proficiency you require isn’t that great. As for the big picture, that’s what management is for. My job is to ensure the network is running. That’s a tough enough job within itself. Perhaps I misunderstood what HP was trying to say. The only cross-silo entity I want to see is the help desk. I have been in environments where you took the various tiers and put them all together under one common manager. Instead of putting all the network people together, you put the support people together, the implementations people together, the engineers together, the architects together, etc. The problem with this approach is that I always needed to interact more with people in my networking silo than I did with people who were in the same tier as me, but may have been storage, server, or security focused. I worked more with people outside of my group than with people within my group. Perhaps other people have different experiences, but from an efficiency standpoint, I favor the silo.
That was just within the first 10 minutes of the HP pitch. I wouldn’t expect to hear much of a difference if another large vendor was presenting. Sprawl is a HUGE problem that things like virtualization have dealt with. What is it about HP that makes them different? Why should you choose them over another vendor when it comes to a networking solution? In HP’s view, there are 3 reasons why.
1. Strong IP in all domains of IT. – You can’t really argue this one. HP has products in just about every major sector of IT. They believe that the only way to present an overall working solution to the customer is to have a fundamental understanding of all things IT. They have a LOT of smart people working for them(as do ALL major vendors) and those people produce a variety of products that make money as well as make our lives easier from a technology standpoint. Check out this link for some proof of that: http://h30507.www3.hp.com/t5/Data-Central/HP-Labs-Releases-2010-Annual-Research-Report/ba-p/88265
2. Open integration – HP continually hammered away at this point throughout their presentations. Everything they do, they want it to be open and standards based. This was their attempt to contrast themselves with Cisco, whom people constantly harp on for all of their proprietary protocols and technology. The problem with preaching the “standards” and “openness” mantra is that you better go to great lengths to ensure there isn’t a hint of anything proprietary in any of your hardware or software. For the most part, HP can make that claim. However, if you dig deep enough, you’ll find that HP has proprietary implementations of certain things. I don’t necessarily think it is that bad of a sin to have some proprietary element to your architecture. Key word being “some”. Juniper is doing it. Cisco, of course, does it. Brocade does it. They all pretty much do it in one form or another. I think you can reach a point to where you are so “standards” focused that you end up like the United Nations. It’s a great idea, but let’s face it. Nobody goes to the UN expecting them to do anything in an expedient and efficient manner.
I will say this about HP’s desire for open integration. They want to meet the needs of their customers in as many areas as reasonably possible. For example, in the realm of storage, HP can integrate with Fiber Channel, iSCSI, and FCoE. In short, they want to give you options.
3. Services approach – Basically, wherever you want to do business, HP will work with you. If you want everything on your local premises, they’ll help out. Outsourced environment? They can help with that too. Even if you are looking at cloud providers, HP can assist with that.
During HP’s presentation, their head of marketing for networking, Jay Mellman, said some things that interested me greatly. Jay said the following, and I am paraphrasing:
“HP has to produce first class technology and HP will never get away with taking second hand infrastructure and slapping it together. Other business lines(server,storage) are counting on HP networking to produce a quality product or they’ll get the product elsewhere.”
Maybe I misunderstood, but the impression I got was that if the networking group produces slop, the other parts of the company won’t use it. In other words, it looks like they only eat their own dog food if it tastes good.
Jay had some more thoughts that he shared with us. He said that it is not about a gold plated network or 100% uptime anymore. As far as customers go, that’s a given. What it is about is the following:
“How do I deliver the right set of services to my customer at a given point in time with the right security at the right cost and then tomorrow morning flip it to a different set of services?”
HP wants to be number 1 in networking. They lead in every other one of their sectors like servers and laptops. They have the marketing know-how and a growing number of people out there who are getting tired of paying Cisco’s premium. The question is, do they have the right technology to pull it off? I’ll leave you with that question to ponder. My next post will focus less on the philosophical marketing stuff and more on the technology that HP is bringing to the table. Stay tuned……
*****Disclaimer: As a delegate for Tech Field Day 5, my flight, food, lodging and transportation expenses were paid for in part by HP. I am under no obligation to write anything regarding HP either good or bad. Anything I choose to write are my opinions, and mine alone. **********
As part of Tech Field Day 5, I received a briefing from Infoblox on their product line. They have some interesting products that revolve around making your life easier in the realm of network services management and network device management. While the products in and of themselves are compelling, the names affiliated with this company are just as interesting.
The VP of Architecture at Infoblox is none other than Cricket Liu. Anyone who has delved into BIND or Microsoft DNS should be familiar with Cricket. I read “DNS and BIND” well over 10 years ago, which Cricket co-authored with Paul Albitz. It’s an industry standard text as far as DNS goes.
In addition to Cricket Liu, another name affiliated with Infoblox, albeit indirectly, is Terry Slattery. Those of us in the network world who keep up with the Cisco CCIE program should be familiar with Terry. He’s CCIE number 1026. Essentially, he’s the first person to pass the lab. CCIE 1025 belongs to Stuart Biggs, who wrote and administered the first CCIE test. The room the first lab was in happened to be numbered 1024. Terry Slattery is the guy who founded Netcordia and created NetMRI. Netcordia was acquired in May of 2010 by Infoblox.
A third name you probably aren’t familiar with is Stuart Bailey. He’s the founder of Infoblox and the CTO. As he himself said during the session with Tech Field Day, he came straight out of academia at the University of Illinois at Chicago and founded Infoblox in 1999.
Infoblox has a fairly straightforward value proposition. Organizations are spending countless hours deploying and administering DNS, DHCP, IP address management, and network configuration/policy management solutions. They aim to solve that with a couple of different products.
First, we have IPAM for Microsoft DNS/DHCP. IPAM is their IP address management product and it does 3 core things:
1) Manage IP address usage. – With a fair amount of eye candy, you can see the status of your entire IP addressing space on your network. By giving you visual maps of IP address usage, you can quickly find the gaps. Need an address allocation of 45 IP’s? You can find a group that large rather easily.
2) Manage Microsoft DNS servers. – IPAM can manage all of your Microsoft DNS servers in a central location.
3) Manage Microsoft DHCP servers. – In a large organization, you might have dozens of DHCP servers. Additionally, you may be concerned about failover capabilities and want to ensure every location has a backup DHCP server provisioned in the event of a failure. IPAM can take care of that for you from a central administrative site.
Second, we have NetMRI. This product came with the acquisition of Netcordia in 2010. NetMRI does what other products like Solarwinds Orion NCM and HP Network Automation software do. It manages the configuration state of your various network devices. With an ability to talk to multiple vendors, there isn’t a lot that NetMRI cannot do. It does several things, but here are the core ones:
1) Archive device configurations. – If you lose a device due to hardware failure, you are probably going to want to put the same configuration on the replacement device. NetMRI can ensure that device configuration backups are done on a regular basis. Any changes made to those devices are logged and over time, you can see what changes were made, who made them, and when they were made. This comes in handy when you need to know specifically when a certain change was made. You won’t always get that from the device itself. Perhaps Juniper devices running JunOS are an exception to the rule as I believe they store a large number of previous configurations on the device. However, if that device is dead, that won’t do you any good unless the configurations are stored on some kind of removable flash memory.
2) Deploy mass changes to devices. – Let’s say your organization has 500 switches on the network and you need to change the NTP settings. Do you want to do that manually? Do you want to build a script to automate that? For most network people, those are not options. There will always be people out there who excel in automation and can write a script in Perl or some other language, extract the device list from a file and make the changes. For the rest of us, you use something like NetMRI.
3) Enforce device policies. – Whether it is firewalls, switches, or routers, you typically have certain things that are always configured on your devices. Some of these are done for security purposes. Others are done for network stability. Imagine that you have a strict requirement for an access list to be applied to all Internet facing interfaces. If someone were to come along and remove that access list from an Internet facing interface, as long as you have a policy configured to enforce that requirement, NetMRI would change the interface configuration back to the way it was before someone changed it. It could then notify you that a policy violation had occurred.
4) Automatic device configuration. – This goes hand in hand with the policy enforcement, but is worth discussing since the benefit here has to do with initial deployment. Imagine a company that has a bunch of remote sites that are relatively similar in nature. Retail, healthcare, and hospitality are a few industries that fit this scenario. If I can simply apply an IP address to a device along with a local user account or SNMP strings, I can have NetMRI do the rest. Why spend time configuring a dozen switches when it can be done through pre-defined policies? How much is that time savings worth to the company?
Infoblox appliances are able to interface with each other in what is known as “Grid Technology”. You can create a small ecosystem of Infoblox products and have them interact with each other. The main focus of the grid appears to be survivability. Multiple appliances can communicate with each other and provide redundancy. If one appliance fails, other appliances in the grid can take over. Every indication I got from the in person sessions as well as research from their documentation leads me to believe that this is strictly related to IPAM. NetMRI can be on a physical or virtual appliance. Although I know it interacts well with IPAM, I don’t think it is a part of the survivable grid.
One final product worth mentioning is IPAM Insight. Although it is designed to map out your network and give you better insight into the connections, one of the side benefits is that it gives you the ability to track down IP addresses and MAC addresses to an individual switch port. I would assume this is a function built into NetMRI, but maybe not. It is built in to some of the competing products. Anyone who has chased down a MAC address that is flapping would instantly see the value in something like this.
What’s the value in all of this?
To be rather simplistic, the value prop from Infoblox is “time”. How much is your engineer’s time worth? Or, to be more brutally honest, how many fewer engineers would you need if you had centralized IP, DNS, DHCP, and network device configuration management? How much is a properly documented network worth?
If you are already in a highly structured environment with defined IP subnets and standard device configurations, you might not see much value in what Infoblox provides. My personal opinion is that no matter the size or state of your network, NetMRI is a solid tool that should be looked at. If you already use one of the competing packages(Solarwinds Orion NCM, HP Network Automation software/Cisco NCM, CiscoWorks LMS, etc) there’s probably not going to be a compelling reason for you to switch to NetMRI. All of those products tend to do the same thing with some minor variations. As for the IP, DNS, and DHCP management, it will only be beneficial in those environments where good practices and documentation do not exist. If your environment is VERY large and you have a million different hands in the pot, IPAM might be a good thing. You’ll be able to lock things down a bit easier, as well as use one central location for administration. If you have everything laid out properly in your Microsoft Active Directory environment, you’ll probably have a hard time selling this to management. The native tools from Microsoft do a decent job of providing usable information. Fortunately for Infoblox, there are tons of those environments that are not managed properly.
Let me know in the comments if you agree, disagree, or need to point out any errors.
*****Disclaimer: As a delegate for Tech Field Day 5, my flight, food, lodging and transportation expenses were paid for in part by Infoblox. I am under no obligation to write anything regarding Infoblox either good or bad. Anything I choose to write are my opinions, and mine alone. **********
I made it back to Nashville before noon on Saturday. A cross country red eye flight with a short layover in Atlanta put me into Nashville just in time. I was able to get a few hours with my kids, dinner with my wife and a bunch of friends from church, followed by dessert and more socializing with all those church friends over at my house. Sunday was full with church, time spent with my father explaining what this San Jose trip was all about(he was very interested in it all), a cub scout hike with my son, and more church. I’m still exhausted. I feel like I haven’t slept in days. I’ve had a nagging cough that air travel made worse and the weather is now 50 degrees warmer than when I left last week to go to California. My co-worker left my company to go work for a well known hardware vendor. His last day was Friday when I was in San Jose. As luck would have it, we had a major data center outage Friday afternoon. I spent the remaining hours in San Jose on the phone and glued to my laptop staring at switch configs. I didn’t get to really say proper goodbyes or even enjoy the final meal with everyone else as I was constantly jumping off and on a conference bridge to deal with the problems in the data center back home. In the end, the problem ended up being something outside of my control, so it was an extra kick in the teeth from the data center gods. In spite of it all, I feel like a million bucks!
Let me tell you why.
1. I love technology. – I love it to the core of my being. There is no greater joy for me than to immerse myself in the 1’s and 0’s of networking and consume mass quantities of information. I’ve never been one to understand people who do what I do for a living and have no real interest in technology outside of 8 to 5 Monday-Friday. Maybe that sounds somewhat elitist. Maybe that’s not a realistic attitude to have. I get paid to learn. That’s the coolest thing in the world. I guess I just recognize that opportunity for what it is and want to be around people who think the same way.
I have been a part of IT groups before where a core group of us had similar attitudes regarding the world of technology. We would feed off of each other and our efficiency and skillsets advanced much faster than all the other environments I have been in where not a whole lot of people shared the same drive and desires. Things change and our careers take us other places. Over time you start to shift back to what is normal for everyone else. You no longer look at Friday afternoon as an inconvenience since you have to put the toys away and go home for 2 days. You no longer wake up Monday morning excited to go into work. For a couple of days last week, I got that spark back.
Now, I don’t want you to think I have a depressing life. I LOVE my life. I love what I do for a living. I love just about everything about my life, and I work in a cubicle! My point, is that I was in the midst of a large group of technology zealots once again. Over the next couple of days, I would either witness or take part in countless discussions regarding networking, storage, virtualization, backups, or systems in general. These were discussions with people who were well versed in their respective areas. People who actually thought about technology as opposed to parroting talking points gleaned from a vendor slide deck. Some of them were published authors. I have a book collecting addiction. Being around authors rates pretty high on my scale of coolness.
2. I love talking to vendors. – My typical exposure to vendors is via their sales channel or third party reseller/integrator. This time, I was able to go straight to the source. I liked the fact that the companies I was exposed to at Tech Field Day 5 ranged from the very large like Symantec and HP, to the very small like Drobo, and Druva. I also saw the companies that fit in between those 2 groups like Xangati, Infoblox, and NetEx. I like talking to the vendors because they all want to differentiate themselves from one another. This means that in general, they have differing points of view as to how to solve a problem. By understanding each vendor’s approach, you can make a more informed decision.
I live on the corporate side of IT. If I make a recommendation in regards to the network, I need to make sure I make the BEST one possible. Yes it takes a lot of time and effort, but choices around hardware and software need to be treated with more care than one uses when selecting which brand of breakfast cereal to buy at the grocery store. I’ll talk to just about any vendor that lives within the network space. No matter how insignificant the product or company may seem, I want to know what it is they do. There is no such thing as being too prepared when it comes to making decisions about your network.
That was Tech Field Day in a nutshell for me. Lots of discussions with my peers and lots of discussions with vendors. For now, I am still trying to digest it all. Two full days worth of briefings and discussions will take a bit to sink in for me. If anything, I have a sincere desire to shore up my virtualization and storage knowledge. I just have to find the time to fit it in. Networking on its own is enough to keep me busy for years to come!
I met some really great and SMART people at this event. Several of them I already knew from Twitter, and some of them I had read their blogs prior to this event. Others were affiliated with vendors, so I had never heard of them, except for some of the people from the larger companies. My RSS feed list has grown by quite a few entries as a result of this trip.
If I could give any advice in regards to this kind of event, it would be this. Go register to be a Gestalt IT Tech Field Day delegate. Do it NOW. If you love technology, if you love talking about technology, and if you want to mix it up with vendors in their own back yard, this is the event for you. I was taken care of very well by Claire and Steven. Nothing was overlooked. Every single vendor that presented seemed interested in us being there. Nothing was off limits in terms of what you could ask. Of course, there’s no guarantee they are going to answer it. The vendors still have to protect their intellectual property and rightfully so. Never in a million years would I have imagined that I would be able to engage someone like the CEO of Symantec and ask a direct question and get a direct answer. I also wouldn’t have imagined myself ever talking to the CEO and CTO of a company like Druva. I spent at least 15 minutes talking with them about their company, social media, and other similar things at the Computer History Museum. Without a doubt it was one of the high points of my trip to San Jose. I could go on and on about other incidents, but it wasn’t my intention to ramble on in this post.
Oh, and lest I forget to tie into the title of this post I should answer the question: “Now what?” Well, I still have to finish preparing to take the CCIE Route/Switch lab. However, I find myself wanting to give equal time to ramping up in the VMware and storage networking worlds. I spent several days in the midst of some storage and virtualization experts. What can I say? They have made me a convert. Or maybe it’s just that I want to understand a bit more of what they were talking about if I ever run into them again. 🙂 In the near future, I want to write a bit about the various vendors. In particular, I will focus on Xangati, HP, Infoblox, and NetEx. They have more of a network-ish focus and that’s the area I focus on. That’s not to say that I won’t comment on the others. I really enjoyed the data deduplication talk from Symantec!
I cannot say thank you enough to everyone who made this event possible. Stephen Foskett played the role of our fearless leader very well. Claire was the driving force behind the scenes making sure everything went off without a hitch. The audio/visual crew produced some very high quality stuff even in the face of several technological glitches. The vendors were very gracious in hosting all of us. I appreciate their interaction from the presentation standpoint as well as their active Twitter presence. Bonus points to Xangati for the bacon and chocolate espresso beans! As for the delegates, well I am humbled to have been among you. Some of you are used to interfacing with these companies at this level. I personally, am not. I do look forward to reading your writings and hope to run into you again at some point!
*****Disclaimer*****
As a Tech Field Day delegate for Gestalt IT, my flights, hotel room, food, and transportation were provided by all of the vendors that presented during this event. This was not provided in exchange for any type of publicity on my part. I am not required to write about any of the presentations or vendors. I received a few “souveniers” from the vendors which were limited to t-shirts, water bottles, pens, flash drives, notepads, and bottle openers.
Image from http://www.save9.com/it-services/telecommunications/cisco-routers-firewalls/
****Note – I am NOT in any way shape or form a VMware expert. I can’t guarantee you that I will be 100% correct in my terminology or representation of VMware, VMotion, VSphere, etc. I apologize in advance. I am just a network guy trying to understand how the Nexus 1000V ties into the VMware ecosystem. I also understand that companies other than VMware are doing virtualization. Please feel free to correct my inaccuracies via the comments.
Paradigm shifts are coming. Some of them are already here. About 5 or 6 years ago I was first introduced to server virtualization in the form of VMware ESX server. For you old mainframe people, you probably weren’t as impressed as I was when I learned about this particular technology.
When it came to VMware, I wasn’t doing anything fancy. I was just using it to host a few Windows servers. When these boxes were physical, they were only using a fraction of their CPU, memory, and disk space. In most cases, they were specific applications that vendors would only support if they were on their own server. From a networking standpoint, there was absolutely nothing fancy that I was doing. All of the traffic from the virtual machines came out of a shared 1 gig port. For me, VMware was a fantastic product in that it allowed me to reduce power, rack space, and cooling requirements.
I realize that some people will take issue with my use of the term “server virtualization”. To some, software and hardware virtualization are different animals. For the purposes of non-VMware people like myself, the fact that I used VMware to reduce the physical server sprawl means that I refer to it as “server virtualization”.
Fast forward to today. It is getting harder and harder to find a company that isn’t doing some sort of server virtualization. It isn’t just about reducing physical server footprint and maximizing CPU and memory resources. These days, you can achieve phenomenal uptime rates due to things like VMotion. For those who are unfamiliar with VMotion, it is a service within VMware that can move a virtual machine from one physical host(ie ESX/ESXi server) to another. This can happen as a result of hardware failure on the physical host itself, additional CPU/memory resource requirements, or other reasons that the VMware administrator deems important.
Today, from a networking standpoint, there are 3 options when it comes to networking inside the VMware vSphere 4 ecosystem:
vNetwork Standard Switch – 1 or more of these standard switches reside on a single ESX host. This would be the vSwitch in older versions of ESX. This is basically a no frills switch. Think of this as managing switches without the use of VTP. You have to touch a lot of these switches if certain VLAN’s reside on multiple ESX hosts.
vNetwork Distributed Switch – 1 or more of these will reside in a “Datacenter”. By “Datacenter”, I am not referring to a physical location. Rather, in VMware lingo, it is a logical grouping of ESX clusters(comprised of ESX hosts). This is the equivalent of running VTP across a network of Cisco switches. You can make changes and have them show up on each ESX host that is part of the “Datacenter”. This particular switch type has several advantages over the standard switch in terms of feature availability. It also allows you to move virtual hosts between multiple servers via vMotion and have the policies associated with that host
Cisco Nexus 1000V – Similar to the distributed switch, except it was built on NX-OS and you can manage it almost like you would any physical Cisco switch. It also has a few more features that the regular VMware distributed switch does not.
That’s the basic overview as I understand it. What I had been struggling with was the actual architecture behind it. How does it work? I can look at a physical switch like the 3750 or 6500 and get a fairly decent understanding of it. Not the level I would like to have, but I understand that vendors like Cisco don’t want to give away their “secret sauce” to everyone that comes along and asks for it.
As luck would have it, my company has purchased several instances of the Nexus 1000V and last week, I was able to spend a day with a Cisco corporate resource and one of the server/storage engineers my company employs. I didn’t realize how deficient I was in the world of VMware until I got into a room with these 2 guys and we started talking through how we would design and implement the Nexus 1000V. I kept asking them to explain things over and over. In the end, a fair amount of pictures on the white board caused the light bulb in my head to go active. I still have much reading to do, but for now I understand it a LOT more than I did. Now, let’s see if I can have it make sense to you. 🙂
The Nexus 1000V is basically comprised of 2 different parts. The VEM and the VSM. If we were to assign these 2 things to actual hardware pieces, the VEM(Virtual Ethernet Module) would be the equivalent of a line card in a switch like the Nexus 7000 or a Catalyst 6500. In essence, this is the data plane. The second piece is the VSM(Virtual Supervisor Module). This is the same as the supervisor module in the Nexus 7000 or Catalyst 6500. As you probably already guessed, this is the control plane piece.
Here’s where it gets a bit crazy. The VSM can support up to 64 VEM’s per 1000V. You can also have a second VSM that operates in standby mode until the active fails. In theory, you have a virtual chassis with 66 slots. In the Nexus 1000V CLI, you can actually type a “show module” and they will all show up. Each ESX host will show up as its own module. Will you ever have 64 VEM’s in a single VSM? Maybe. However, there are limitations around the Nexus 1000V that make that unlikely.
The VEM lives on each ESX server, but where does the VSM reside? It resides in its own guest VM. You actually create a separate virtual machine for the VSM when installing the Nexus 1000V. That guest VM resides on one of the ESX servers within the “datacenter” that the Nexus 1000V controls. You access that guest VM just like you would a physical switch in your network by using the CLI. Once the VSM is installed, the network resource can go in via SSH or Telnet and configure away.
That’s the basic components of the Nexus 1000V. There are other things that need to be mentioned such as how communication happens from the guest VM perspective to the rest of the network and vice versa. Additionally, we need to discuss the benefits of using the Nexus 1000V over the standard VMware distributed switch. There’s a lot more than just the management aspect of it. I will cover that in part 2. Additionally, I plan on doing a write up on the Nexus 1010 appliance. This allows you to REALLY move the control plane piece out of the VMware environment and put it on a box with a Cisco logo on it.
He that outlives this day, and comes safe home, Will stand a tip-toe when this day is nam’d, And rouse him at the name of Crispian. He that shall live this day, and see old age, Will yearly on the vigil feast his neighbours, And say ‘To-morrow is Saint Crispian.’ Then will he strip his sleeve and show his scars, And say ‘These wounds I had on Crispian’s day.’ Old men forget; yet all shall be forgot, But he’ll remember, with advantages, What feats he did that day. Then shall our names, Familiar in his mouth as household words- Harry the King, Bedford and Exeter, Warwick and Talbot, Salisbury and Gloucester- Be in their flowing cups freshly rememb’red. This story shall the good man teach his son; And Crispin Crispian shall ne’er go by, From this day to the ending of the world, But we in it shall be remembered- We few, we happy few, we band of brothers; For he to-day that sheds his blood with me Shall be my brother; be he ne’er so vile, This day shall gentle his condition; And gentlemen in England now-a-bed Shall think themselves accurs’d they were not here, And hold their manhoods cheap whiles any speaks That fought with us upon Saint Crispin’s day.
Such are the words that William Shakespeare penned in “Henry the Fifth”. They come from the Saint Crispin’s Day Speech that King Henry V gave prior to the battle of Agincourt in 1415 where the English defeated the French and King Henry ended up with a French princess named Catherine as one of the spoils of war. Although the speech from Shakespeare is made up, it is still a beautiful combination of words that express the pride the English soldiers would feel in the years after the battle. Others might forget what went on, but the soldiers would never forget. They would be a part of history. Which brings me to the point of this post……
“regardless of your thoughts on IPv6 adoption, it’s a pretty interesting time to be a networker”
That’s putting it mildly and it got me thinking about the changes going on in networking these days.
1. IPv6 Transition – Certainly you have heard of IPv6 and the coming IPv4 address exhaustion. If not, you need to get out more.
2. Virtual Networking – With the explosion of vmWare and other virtualization vendors in the past several years, a fair amount of traffic is cruising around “virtual” switches inside physical servers. Guess what? You still have to manage it. You still have to secure it.
3. Wireless Explosion – Everything is wireless today. Cameras, printers, phones, tablets, laptops, and other wireless capable devices are growing in number each year. If you aren’t familiar with wireless, you better be soon.
There’s more. Storage traffic riding over the same wire as voice, video, and data. How about link encryption on your internal switch/router infrastructure? Don’t forget the rush to flatten datacenter networks to L2 courtesy of TRILL or each vendor’s implementation of it.
Some difficult and interesting days lie ahead. Difficult and interesting from the standpoint that we’ll have to implement things that we haven’t been doing for years and years. This is new ground for many of us. With the right amount of due diligence and a couple of heavily padded blocks of time from various consultants, it will all get done. Fast forward to a few years down the road. Like King Henry said in Henry V:
Then will he strip his sleeve and show his scars, And say ‘These wounds I had on Crispian’s day.’ Old men forget; yet all shall be forgot, But he’ll remember, with advantages, What feats he did that day.
and
And gentlemen in England now-a-bed Shall think themselves accurs’d they were not here, And hold their manhoods cheap whiles any speaks That fought with us upon Saint Crispin’s day.
We’ll all have scars, but they’ll be scars we can be proud of. This is an interesting time to be in networking, but I wouldn’t want it any other way. I foresee changes like these flushing out people who are not ready for the paradigm shifts that are coming or are already here.
New blood will come into the field. You’ll be able to guide them and mentor them and show them your scars from the IPv4, every server was physical, no wireless, TDM PBX, Frame Relay was king days. Then, they’ll produce a fake smile as you bore them with stories of how many CAT5 patch cables you have made in your past career and then they’ll mock you when you’re not around. Kind of like how we mock Thomas Watson and his inability to predict the demand for computers.
Posted incareer, learning|Comments Off on Be A Part Of History
Cisco has killed the Flip. Twitter and blogs are replicating this announcement out to the far corners of the Earth. I am cheering as are plenty of other network centric people.
