Plug-n-WiFi December 28, 2009
Posted by Devin Akin in : Uncategorized , 2commentsI know this entire blog is wishful thinking at this point in time, but…here goes anyway.
Think Vista vs. OSX. Just by reading the title, you know what I’m saying… (BTW, Mr. Jobs, feel free to send me a 13″ MacBook Pro, loaded to the gills, for the free marketing).
Is it just me, or is anyone else sick to death of all of the complexities of enterprise Wi-Fi? I’ve been in this market since…well…before it was really a market, and it’s always been the same old story: more functionality means more complexity…and that functionality just keeps on coming. Why can’t it just be Plug-n-WiFi? Client connectivity and troubleshooting, gathering network-wide stastical data, visualizing and automatically reacting to the RF environment (or better yet, acting on predictive data), and deploying APs in the right locations all need to be very automagic. Heck, I’d be OK with voice-activated Wi-Fi configuration. 
By the way, I think I’ve found the problem…
Most people in Silicon Valley are far too geeky. There are countless brilliant engineers there with IQs that could be used to describe the temperature on Venus, and they drive the boat. These engineers play roles like Product Management, Engineering, Marketing, and yes, even CEO. How then are we ever going to reach a point where someone is relating to normal people (like me)? For this blog, I define normal as “not as smart as most of those engineers I just described.”
There are a certain number of tasks that seem unavoidable, such as creating profiles (a set of parameters assigned to a user group) and hooking the Wi-Fi infrastructure into Active Directory and other backend systems, but even those should be Minority Report simple.
You know, mesh is a big piece of this too because cabling is a major pain and cost. Yeah, yeah, I know, some of you overly-analytical types are thinking, “well, the cable is already there, so you might as well use it.” To that, I say, “in the future, it would be nice not to have to worry about cable plants.” I have a nice, crisp five dollar bill (picture me snapping it sharply in front of you) that says there will come a time when Cat5/5e/6 won’t do the trick, and new cable plants will be required.
Today’s SOHO gear is almost Plug-n-WiFi, when it supports Wi-Fi Protected Setup (WPS) of course. It would be super cool to have an “Insta-Fi” enterprise system. You push a button, and voila – Wi-Fi. You want mesh? Hit the ON button and…Wi-Fi. Did I mention that this blog was about wishful thinking? Hooking the Wi-Fi infrastructure to the Wi-Fi management system should be Plug-n-WiFi too. Power on APs, and…done. You know what I think we have to do to reach this goal? We need to hire dumber CEOs (or at least CEOs who don’t have an engineering background), dumber Marketing VPs, and dumber Product Managers. Why not dumber Engineers too? Because somebody has to build all of the easy-to-use stuff the dumber people come up with! I have supreme confidence that engineering can build anything you ask for…but the problem is that everyone is asking for the wrong thing. These smart people are killing us!
Besides being mega cool for the enterprise, my pipe dream also has back-end advantages for the Wi-Fi infrastructure manufacturer. Think about the cost of field sales teams. Salaries, travel, entertainment, and other expenses…in order to win deals and to support VARs. If my Plug-n-WiFi dream existed anywhere in today’s market, it would mean that inside sales could handle 90% of the sales process. Good inside sales guys would rake in the dough by selling, “Enterprise Wi-Fi by dumb guys for dumb guys.” There wouldn’t be anything complicated to worry about, so those super-smart field sales teams would be wondering what to do with themselves. 
Of course, you know that as soon as I publish this blog, System Engineers and Account Managers are going to start blasting me about their jobs. Hey fellas, don’t worry, nobody in the industry has realized my pipedream yet, so you can keep your job for a while longer. 
What can I say? You got screwed by being born so smart.
/Interlude-start
At Aerohive, we have the dream team of SEs. Holy smokes batman, these guys are awesome.
/Interlude-end
So where do we start (besides hiring dumber people in non-engineering roles)? I think the first thing is to understand how users think and then build a GUI interface that is so intuitive, that very little, if any, thinking is required to use it. All vendors need to come to a consensus on what a GUI interface needs to look and feel like. Perhaps the Wi-Fi Alliance needs to drive this effort. They could call it “Task Group Easy Button.” Next, we need to turn what is currently a manual transmission into a 6-speed automatic. Features, like Active Directory integration, that cause SEs to break out in a cold sweat, should be “PhD” simple – that is, “Push here Dummy.”
A conversation between you and your Wireless Network Management Systems (WNMS) might currently go something like tihs:
You: HAL, can you please integrate with Active Directory?
You: Well what about applying an Application Layer Gateway for TFTP to SSID Engineering?
You: OK, well if you’re not going to do that, how about changing the firewall settings on SSID Sales to block incoming Telnet traffic?
You: Dude, what are you good for anyway? Grrrr. (Clanging and banging heard in the background…)
I’d love to hear some feedback on this one.
Manual or Automatic? December 21, 2009
Posted by Devin Akin in : Uncategorized , add a commentMuch of today’s Wi-Fi troubleshooting is a manual process, using one kind of analyzer or another. If you want to see layer-1, you bust out your Cognio or Wi-Spy, and if you want to see layer-2, you crank up OmniPeek, CommView/WiFi, AirMagnet Wi-Fi Analyzer (or AirMedic), or Wireshark. If you need a little more of the big picture, you might reach for AirMagnet Survey or Ekahau Site Survey. Am I wrong? So that brings me to the point of this article…
Will automated, infrastructure-side diagnostic tools ever completely replace manual troubleshooting tools? I think it will follow the same path as manual and automatic transmissions in the automotive industry. Everything started out manual, and then the market slowly progressed toward automatic transmissions until automatics commanded the lion’s share of the market. If it weren’t for those few hold-outs (sports cars, large trucks, and motorcycles), manual transmissions would be a thing of the past. Thank goodness for those few hold-outs! Consider that sports cars, large trucks, and motorcycles all do extreme tasks, and the run-of-the-mill stuff all uses automatic transmissions. I think it will go the same way with Wi-Fi troubleshooting and performance optimization.
Good protocol analysts have always been the creme-de-la-creme of IT (wired or wireless), and so there are only a few of them (relatively speaking). In the future, there will be an even smaller normal-to-analyst ratio, and they will be the “race car mechanics” of the industry – tweeking for high performance and finding needle-in-a-haystack problems. Analysts won’t go away completely, but the protocol will soon be so large and sophisticated that only a “few good men (and women)” will be willing to engage at a level that will allow for excellent troubleshooting skills.
A problem I forsee is that it will become more difficult for companies who make analysis software to make enough money on the products to justify making them at all. Smaller “one trick pony” companies will get swallowed up by larger companies where their analysis product(s) will be among a larger portfolio. That will help some of course. If manual analysis products are left as stand-alone products, this conundrum may mean lower quality products because it’s hard to justify the time and resources to work on them due to a small quantity of sales. It’s more likely that analysis products will integrate with (already happening) or be integrated into Wi-Fi infrastructure as yet another diagnostic tool (albeit a really good one).
It’s really quite staggering at all the cool diagnostic tools in the infrastructure now. One of my favorites is the ability to stream 802.11 frames from APs directly into a desktop-based protocol analyzer over an Ethernet connection. That’s one that won’t go away easily I’ll bet. Once vendors have a comprehensive set of diagnostic tools in the infrastructure, it’s only a matter of time before they automate those tools in such a way that the infrastructure can self-manage security or performance problems, guaranteeing a given level of performance. That will be the beginning of deterministic performance…of Wi-Fi as a utility. It’s like having a 400hp V8 with a 6-speed automatic. 
I’m interested to hear what you think. Is my crystal ball out of kilter or did I hit the mark?
Skating To Where The Puck Will Be December 14, 2009
Posted by Devin Akin in : Uncategorized , 2commentsIt’s not breaking news that controllers are the de facto standard in today’s market. Most folks looking to buy sizable enterprise Wi-Fi infrastructure are looking for the biggest, fastest, meanest controller they can find. Just talking about it makes them grunt like Tim “the tool man” Taylor. I’m no stranger to that feeling…
Before I jump to any early conclusions, let’s talk about where the Wi-Fi infrastructure market is headed. That sounds reasonable, right? So, let’s start at the beginning…
Once Wi-Fi is installed, there are some undeniable facts.
1) Users and administrators will eventually ask for more bandwidth.
2) Users expect a wire-like application experience.
3) Once Wi-Fi, always Wi-Fi. There’s no going back. There’s no giving up the flexibility and ROI.
Certainly these are not all of Wi-Fi’s undeniable facts, but rather just the ones that came to mind while writing this blog post. Now on to logical Wi-Fi infrastructure progression…
To handle the rising onslaught of user demands, controllers must get much larger (physically or virtually), more numerous, and eventually both. The progression looks like this:
1) Bigger controller
2) Another big controller
3) Virtualize all big controllers into one giant controller
Then you have a choice to make…
A) Exponentially increase the synaptic connection (interface speed) between the giant Virtual Controller and the Ethernet core
B) Merge the giant virtual wireless controller into the Ethernet core, forming one giant monstrosity of a network core with 100 MB firmware files and 5 MB config files.
Either way, you then realize that you’re still not leveraging your Distribution and/or Access Layer Ethernet network…so here is, like Paul Harvey says, “the rest of the story.”
4) Introduce Distribution and Access layer controllers to segment user traffic and to push the traffic u-turn closer to the wireless client. If you took step B above, then it’s entirely possible that you now use Wi-Fi code in your Distribution and Access layer Ethernet switches.
5) Sudden realization that the network design is a serious mess.
More-or-less, you’ve now built a core overlay, a distribution overlay, and an access overlay – all in order to get the intelligence closer to the edge for scalability. Why? You could skip all of that nonsense and just push all of the intelligence (forwarding, filtering, QoS, security, locationing, RRM, etc.) down into the AP to start with and have the APs coordinate among themselves. This leverages your existing Ethernet infrastructure, eliminates introduction of a parallel “overlay” network due to controllers, and eliminates single points of failure and bottlenecks. All roads lead to Rome. The question is how long is it going to take manufacturers to get there? The controller architecture was a half-decade detour due to a lack of affordable AP hardware, but that era is running out of paved road.
If you think about it, the only thing wrong with autonomous APs was the fact that the control plane lived in each AP individually rather than being spread across multiple APs. That, in fact, was the sole reason for a controller in the first place. By making one simple change, you bypass an entire generation of high-cost Wi-Fi infrastructure plagued with bottlenecks, latency, and single points of failure. This “simple change” obviously requires more powerful APs, but with today’s commodity hardware pricing, that’s not a big deal.
If you’re an Apple fan, you know the kind of anticipation that builds until the time they actually release a new product. There are myriad rumors about what the product will look like, what it will be called, its specifications, etc. You’re holding your money hoping to make the big score just as the new Apple widget hits the market. For months you’ve told yourself, “it’ll be worth the wait!” I know that feeling. You’re “skating to where the puck will be.” Some folks treat Wi-Fi infrastructure in the same way, and they’ve been holding onto their autonomous APs, “power-braking” for lack of better words, waiting for a next-generation product that will allow them to bypass the detour-du-jour and get long-lasting bang for their buck.
Until next time, I’ll leave you with this little blurb from The Matrix. Hopefully you will figure out how it applies to this blog post.
“Were you looking at me Neo, or were you looking at the woman in the red dress? Look again.” –Morpheus
The Bloginning December 10, 2009
Posted by Devin Akin in : Uncategorized , 9commentsAnyone who has followed my blog very long knows that I speak my mind – sometimes to my own detriment. I assure you that readers will get more of the same with this blog. I’m fortunate to be a part of a management team who understands the importance of a company connecting with its audience (partners, customers, media/analysts, and more) and telling the truth – plainly. If you know me at all, you know how much I hate vendor Kool-Aid, and that hasn’t changed one bit. My agreement with Dave Flynn (our CEO, my boss) when coming on board at Aerohive was that if I felt like it was kool-aid, I didn’t have to drink it or spew it. So, when I blog, speak, or write, I will only write the things I believe are true. If you can prove me wrong, I’m happy to change what I said. If it smells like kool-aid (marketing spin that is “the best version of the truth”) to me, I’ll let someone else say it.
Since this is my first blog post, I want to briefly speak to why I left CWNP, a company that I co-founded with Kevin Sandlin in 1999, to join Aerohive Networks. CWNP was and still is a very important part of the Wi-Fi industry, crucial to educating and certifying industry professionals, but after 10 years of doing very much the same kinds of things (writing exams, books, whitepapers, and more, teaching, blogging, speaking, consulting, lab testing, etc), I wanted to get more involved in “the game.” Because I spent so much time testing vendor hardware and software and working with various manufacturers’ engineering, product management, and QA departments, I got to look under the hood in a variety of ways. It was a fun (though exhausting) experience to be sure.
The trick for me was that if I were going to work for a Wi-Fi manufacturer, I would certainly be giving up 10 years of vendor neutrality and a position as an industry analyst (of sorts). Deep down, I was thinking, “if I pick the right horse, the result will be worth more than what I had to give up to get it.” I’m a risk taker, so I set out to find the right horse.
Don’t worry though, I wouldn’t leave you stranded… There are other highly-capable and experienced minds in the Wi-Fi market to keep you abreast of things from a vendor-neutral perspective. You probably already know who they are…Paul DeBeasi, Joanie Wexler, Lisa Phifer, Dave Molta, Tim Zimmerman, Craig Mathias, John Cox, and other industry spokespeople on whose words we can depend. Lucky for us, the industry has no shortage of professionals that will give it to you straight.
So how was I supposed to define, “the right horse?” I tried to keep it simple really. I prioritized the things that matter most to me:
1) Integrity. Does the company tell its partners, customers, and the media the truth even when the truth doesn’t sound very good? I’m not talking about, “the best version of the truth”, but rather, “the bare naked truth.”
2) Corporate Culture. Can I be myself and do what I do best among a group of like-minded professionals who care about each other and work as a team? Is everyone treated with respect, yeah, even love? Yes, seriously.
3) Technology. Is the company creating a platform that’s based on solving tomorrow’s infrastructure problems? What is its technical and cost model differentiation? What has been their success to-date?
As you can imagine, there are additional tidbits here and there that had to make sense for me, but these were the biggies. I would go into how Aerohive was the best in these areas (in my humble opinion of course), but that’s evident by the fact that I’m working here. I would like to thank the other vendors for their kind and generous offers. It was very much appreciated.
Now that I’m here, and have resurfaced after dropping off the radar for a couple of months, it’s time to tell you what I’ve been up to. Part of my job is customer advocacy. It consists of making sure that customers (whether end-users, partners, or employees) are happy and the company is meeting their needs. Other than advocacy, I’m more-or-less “the cleaner.” Have you ever seen the movie “Michael Clayton”? I’m a cross-departmental “doer of stuff” that equates to sales, marketing, engineering, operations, and products. It’s the most fun I’ve ever had, and I’m more efficient…and exhausted…than I’ve ever been.
Aerohive intends to set a new, and significantly higher set of customer-impacting standards in the Wi-Fi market. We intend to do this with the best team, the best technology, the highest standard of ethics, and the most customer-centric initiatives the market has ever experienced. For you who have followed me since the humble beginnings of CWNP, you know that I make good use of my mild case of OCD, and I have a great team here to help me continue the tradition.
Currently I’m traveling all over creation speaking to groups, partners, and customers – occasionally slipping in a webinar here and there. I’m doing a bit of writing as well. Perhaps you’ve read some of Aerohive’s latest whitepapers and solution briefs. 
I now have the privilege of working with some of the most innovative people I’ve ever met (including Bob)…and you know the Bob I’m talking about.
There’s lots of work to do toward Aerohive’s new-and-improved goals, and, like the Marines, we’re always looking for a few good men (and women). Simpli-Fi.
