Aruba Model 220 Access Point new with 802.11ac!

This week was the official launch of Aruba's new 802.11ac access point. Blake Krone, Chris Lyttle, Daniel Cybulskie, Keith Parsons, Ryan Adzima and I attended the product launch announcement as a members of the Tech Field Day Roundtable group. The day's agenda for the kickoff event was to announce the technical aspects of the new access point, perform a live demonstration of its capabilities and then we met with key people from Microsoft, Netflix and Exafort to learn more about how Aruba has strategically partnered with vendors and customers to fine tune Aruba's own products and provide advanced support for enterprise customers.

Aruba 802.11ac Announcement with Keerti Melkote, Aruba CTO and Founder

Keerti Melkote took us through the history of Aruba Networks, describing the product features released since the company founding. He spoke about the Meridian acquisition. The acquisition of Meridian provides location based information, the ability interacting with exhibits in museums through the Meridian application, and push notifications to client devices.

The future goals of wireless networks are to enable the all wireless office

• Move to 802.11ac
• Aruba 220 series AP with ClientMatch
• Unplug the desk phone
• AppRF with Microsoft Lync Visibility
• See the apps and the air
• ArubaOS 6.3 and AirWave 7.7

The new Aruba 220 802.11ac access point has a $1295 list price, and is available as a controller managed and controller-less. 

• Lifetime warranty
• 3x3:3 dual radio (turbo QAM)
• 2X GE link aggregation
• 1Gbps TCP throughput
• High availability 
• Operates with 802.3af, requires 802.3at for full functionality.
• Purpose built instead of modular
• Power draw is 15w vs 25w

The enclosure design of the access point now has an iso plane to separate the antennas from the board as well as no ventilation holes. This solves the problem of not being able to install other Aruba access points with ventilation holes in environments where the access point might be exposed to moisture and dust.

Peter Lane then demonstrated the performance of the Aruba AP 220 802.11ac access point through a live demonstration.

Aruba's test results achieved 830 Mbps on a dual stream laptop, and 240 Mbps on a single stream smart phone.

The fastest 802.11ac rates drop down to 802.11n speeds at 50 ft from the access point. Maintaining this desired distance from the access point is how you can deliver the capacity of 802.11ac to mobile client devices.

The advancements that Aruba has made in ClientMatch allows this technology to work at several different layers of the OSI protocol stack.

• ClientMatch (L1) for link optimization moves clients to APs that have better signal strengths for them to connect to. Simply link optimization for client devices.
• ClientMatch (L2-L3) for traffic optimization takes AP load into consideration when moving clients to other APs. Uses signal strength and load of the AP.
• ClientMatch for App Optimization (L4-L7) is a version of SDN to optimize the WLAN to the client/app information.

The client testing Aruba has performed uncovered that the Android software version 2.3 roams notoriously badly.

ClientMatch has device type identification built into it and helps ClientMatch make moving decisions. ClientMatch will stop steering clients for 2 months at a time to stop causing problems with a client device that doesn't want to roam. Aruba is optimizing the bottom 20% of the network instead of the top performing devices that could be improved a small percentage more.

VisualRF now shows the health of the client by the colored circle shown around the client device. Visual RF now integrates with Lync API via a diagnostics API. The QoS for all Lync apps, call admission control and Wi-Fi call quality stats can be shown through Visual RF. Aruba Networks is the only Lync qualified wireless LAN.

The visibility into the diagnostics API of Lync is new. Soon Aruba will be able to tie in Lync diagnostic information into ClientMatch to make more intelligent decisions on client roaming based on Lync application data.

Future of wireless networks

• 2013+
• Mobile
• Personal LANs
• Software-centric
• L4-7 based
• Open architecture

Microsoft Lync over Wi-Fi with Pascal Menezes, Sr. Program Manager at Microsoft

Pascal presented the history of the partnership between Lync and Aruba Networks.

Lync is the enterprise version of Skype. Lync can be an instant messaging platform which can determine presence across Lync and Skype platforms.

Lync 2013 Mobile Client Features don't support viewing shared Lync content on anything other than an iPad or a Windows 8 or Windows RT OS. I found this interesting since it was the only features line without all the check boxes checked. 

Microsoft started a Lync Wi-Fi Partner Program and Aruba was the first partner to work with them. The qualification program means Microsoft has done all the testing required to ensure interoperability with vendors' wifi networks. The Lync Network Diagnostic API is the forward looking portion towards Software Defined Networking (SDN) in wireless networks.

Aruba OS 6.1.3.2 and higher is the firmware version Lync tested.

The Lync/Aruba interoperability testing environment

• Two buildings were upgraded to support real-time media.
• Testing facility was B30 and B31 in Redmond. Each building has 600 Lync users.
• There are a total of 182 Aruba APs (about every 60 feet apart). Based on the trial results, MSIT is upgrading the entire global Wi-Fi infrastructure to 802.11n.
• Spent two years tweaking the wireless network.
• Signals -45Bm to -65dbn and snr better than 30db
• OKC enabled (no 802.11r) fast BSS transition support
• Enabled ARM
• WPA2 in enterprise mode
• Adjusted the DTIM timer to 3 for mobile devices to save on battery life
• The sticky clients they saw would hang at -90dbm and then roam to an AP with -60dbm.
• Client devices would experience seconds of audio outages during these roam times.

The rate adaption algorithms and TX retries may take up too much airtime in retries for UC real-time media traffic. Existing STAs rate adaption algorithms not well suited for UC.

The Microsoft Lync team has proposed for a Mobile Multimedia Over Wi-Fi WFA WG

Support of 18 Wi-Fi vendors. The goal of the proposal is to develop a certification that improves the Wi-Fi roaming and application performance in enterprise and public venues in a manner that is impelling for vendors. The ability to verify performance for real-time voice and video over Wi-Fi is the ultimate goal of this proposal, as QoS is difficult to deploy and is expensive and complex to manage. In most enterprise networks, QoS is not widely implemented end to end (from the wireless to the wired networks).

After the presentation from Pascal, we had our roundtable discussion about the Aruba announcement and our opinions on 802.11ac. We discussed sticky clients, antenna design of the Aruba 220 access point, the Lync diagnostic API and our wish for a portable tool which would allow wireless engineers to view the Lync diagnostic data without requiring a login to the Lync server.

Designing Wi-Fi for Voice & Video with Mike Kail, Netflix VP of IT

Mike D. Kail VP of IT Operations at Netflix described how 802.11ac affects their all-wireless office environment.

[link to video will be added later]

Netflix uses Wi-Fi for Voice and Video throughout their offices. Netflix has a 100% mobile office. This increases productivity to be able to work anywhere within wifi range. Some graphic designers have iMacs that are hardwired, but everyone else is wireless. The docking station may be used for Accounting people, but it's rare to be wired into the network. Netflix employees still have desk phones but most people use their cell phones. They are desk phone optional (internal recruiting team and the legal team probably use wired phones).

The ultimate goal of the Netflix network design is to achieve a Zero Trust Network

• Goal is no perimeter firewall 'gate' 
• Identity is the new perimeter
• Moving all devices to EAP-TLS
• Evaluating ClearPass
• They use Google+ hangouts for collaboration
• Telepresence and Lifesize are used for video conferencing.

Netflix prioritizes multimedia on the same SSID as data and all Android and iOS devices are allowed, no exceptions. Netflix is beta testing Aruba's 802.11ac access points and evaluating Aruba RAPs for remote locations to extend corporate networks to home users.

Netflix Content Operations does the QC streaming testing of the UI across wifi networks. The Los Gatos campus is 5 buildings with 280 APs, and at any given time, there are 1300 - 1400 people accessing the Internet. Typical Netflix upload/download speeds are 199 Mbps down / 174 Mbps up.

Netflix employees can use any device, as they've implemented security to access to the data, not the access to the device.

Netflix had some original access point placement issues, some areas have multiple SSIDs and there are a few roaming problems that are problems on the client side. Overall, Netflix has one main SSID and a couple others that give them presence into other countries. The additional SSIDs are locked down to specific floors for testing.

Smart TVs and Apple TVs (due to the authentication method used [EAP-TLS]) are wired.

100% EAP-TLS, no captive portals, but there is an open guest network that uses ClearPass. Netflix is also looking to implement their own open source certificate framework. Netflix uses CACTI and MRTG to monitor their wireless network in conjunction with AirWave. Netflix is also active in Github and uses open source software whenever possible. Netflix is also working to implement NAC on the wired ports to form the Zero Trust Network.

Next-gen Access Network Design with Arun Kanchi, Exafort CEO

Arun Kanchi CEO of Exafort

[link to video will be added later]

Eric Pasteur

Mark Johnson

Exafort is a Cloud and Mobile Systems Integrator

Mobility needs summarized in three words:

• Security
• the security of corporate data
• Privacy
•  the personal privacy of workspace apps
• Experience
•   ease of use, self registration, content based access

Exafort manages the wireless infrastructure for Arista for site surveys, design and deployment as well as IT help desk as a service. Arista's IT infrastructure is 90% cloud based - Wi-Fi, tools, corporate business applications. Exafort has planned an AP deployment ratio of 75 access points to 1000 people. Exafort has planned a 100% redundant wireless LAN design. They've planned the access point placements so that there was bleed over between floors and this bleed through coverage would support users on that floor.

Exafort is pulling two cables to the APs, the additional expense is the physical cabling. They're not having to pull a new run. The second run is half the cost of the first run. The ports on the AP will only have one live at a time, but each access point is occupying two full switch ports. The new Arista building is going live in the fall of 2013 and will be the first wireless network for Arista. Exafort has been managing Arista's IT infrastructure since 2009.

Exafort used AirWave to do a predictive survey for the new building deployment. Exafort didn't use SNR or RSSI metrics within AirWave to design the predictive deployment, the used the coverage model to design so that all areas of the building were covered in signal strength shown in green in AirWave. Exafort is using a 20 MHz channel plan for 2.4GHz and 40 MHz for the 5 GHz, and they're relying on RRM to auto adjust the channels/and keep the noise floor low.

To wrap up: I was very excited to be invited to attend this launch in person! There are so many things happening in the wireless space: product launches, acquisitions, changes in RF design due to high density clients and the continual advancements of consumer devices. It was an honor to be invited and contributed my two cents worth to the discussion. Thank you to Aruba Networks and Gestalt IT for inviting me!

Smart Meters: Now With More Data and Less FUD

Recently I read a news article about two mothers who were arrested on their property as a result of them blocking access to their utility meters in order to prevent the utility company from replacing their analogue utility meter with a digital "smart meter". The mothers are members of the Naperville Smart Meter Awareness group (NSMA), and NSMA has a federal lawsuit pending against the City of Naperville regarding the installation of the smart meters in the city. NSMA is concerned that the smart meters will affect health, security and privacy.

NSMA also objects to the Naperville Department of Public Utilities referencing a document created by the California Council on Science and Technology (CCST) titled "Health Impacts of RadioFrequency from Smart Meters" as evidence that smart meters pose no threat to public health. They claim that the CCST document has numerous flaws and does not address whether the current FCC guidelines are sufficiently protective of health considering current levels of RF exposure (which includes ambient levels from other devices), and the cumulative effect over time. The NSMA group also cites nuclear physicist Daniel Hirsch's commentary on the CCST data as an indicator that the smart meter radiation levels could be 100 times more exposure than cell phones.

I began to do some research, starting with looking up the FCC ID of our smart meter [FCC ID: R7PER1R1S4]. The information I found states it is made by Landis+Gyr and is called the Focus-2. The data that I found on the L&G Focus-2 unit via the FCC testing data states the average RF output power is ~16dBm (which is equivalent to 40mW) when operating in the 902.1-927.9MHz frequencies. The L&G Focus-2 UtiliNet endpoint has an integrated loop antenna, located on the PCB surface layer on the reverse (non-component) side of the assembly. According to the FCC test report, the antenna has a typical gain of ~3dBi.
From the UtiliNet Endpoint User Guide:

UtiliNet is a comprehensive wireless data communications solution that utilizes spread-spectrum radios in the 902-928 MHz area of the radio spectrum to provide reliable network answers for remote telemetry or distributed control applications. UtiliNet radios combine three important technologies: a mesh architecture for peer-to-peer communications and true networking functionality, asynchronous spread spectrum frequency hopping for maximum use of bandwidth, and packet switching for guaranteed message transfer and automatic store-and-forward routing. 

The communication language/protocol used between the UtiliNet smart meters (mesh nodes) is the Device Control Word (DCW) language. Typical application information requests are radio configurations, radio queries, data collection, communication to end devices, protocol translation and peer-to-peer control.

I did my own research to verify the data for whole body SAR being used in these reports. Sometimes the SAR value was expressed in microwatts (μW/cm2), other times it was expressed in watts (W/cm2). Luckily there's a couple of handy websites that'll do the conversion for you. Microwatts to Watts and Watts to Microwatts.

The CCST report contains the following chart showing the values they've used to calculate the threat level of the smart meter radiation output:

The report by Daniel Hirsch commenting on the CCST report contains the following chart showing the numbers he's used to calculate the threat level of smart meter radiation output.

I have a few issues with the wording and the data used in Mr. Hirsch's report. These following two paragraphs precede the chart above in his report and the wording used strongly suggests his chart is not using definitive values.

It is strongly recommended that CCST revise its Draft Report and conduct actual measurements of cell phone, microwave oven, and SmartMeter RF cumulative whole body power densities. If measurements aren’t made, then rigorous calculations correcting for cell phone and microwave oven duty cycles and whole body exposures should be made. 

A summary figure below shows how rough estimates of the effect of those corrections suggest SmartMeters may produce cumulative whole body exposures far higher than that of cell phones or microwave ovens. 

His repeated use of qualifying words (summary, rough, suggest, may) gives the impression that he is not making a definitive conclusion with the data being displayed in the chart. The problem I have with this is that Mr. Hirsch's chart is being used as factual evidence to the risk of exposure to smart meters.

I took it upon myself to find neutral Internet sources documenting the radiation output for microwaves, the L&G Focus-2 smart meter (since this is the one installed outside my residence), the iPhone 5, Samsung GT-I9500 and the FCC limits for RF exposure. Below is the chart I've compiled presented on the left in microwatts and on the right in watts.

Here are the links to the sources I used for the data represented in my chart shown above:

iPhone 5 radiation testing 1.18W/cm2
Samsung GT-I9500 radiation testing 1.55W/cm2
L&G Focus-2 smart meter .000018W/cm2
Maximum microwave oven leakage data .0000005W/cm2
FCC limits for RF exposure (FCC 13-39 3/29/2013) .08W/cm2 - 1.6W/cm2

Mr. Hirsch also expressed concern that the duty cycle of the smart meters wasn't represented  accurately in the CCST report. His report assumes a 100% transmit duty cycle for the smart meter output (that is to say that the smart meter is transmitting/generating signal 100% of the time).

As I began to read about the L&G Focus-2 smart meter installed where I live, I realized I could use the Metageek WiSpy 900x to view the frequencies the smart meter operates (902-928 MHz). I made a few recordings indoors and outdoors in hopes of picking up the transmissions from the smart meter. I wanted to see if I could ascertain how often the smart meter was transmitting, and for how long. In this video clip, I start off with the outdoors recording. I was positioned at a distance of 6 inches from the plastic cover to the smart meter. There were intermittent bursts of RF energy detected, but there did not seem to be a predictable, repeatable pattern to the transmissions. The second half of the video is the recording that was taken indoors, approximately 10 feet from the wall where the smart meter is installed. The RF energy detected indoors was not as strong as the signals picked up when positioned closer to the smart meter.

The RF energy bursts detected by the Metageek Wi-Spy 900x were very brief and did not use up a large portion of the available spectrum. Duty cycle utilization was minimal, given the short transmission windows. It is clear that the L&G Focus-2 is continually powered on with electricity, but that the smart meter is not transmitting continuously, nor is it utilizing a 100% duty cycle.

The videos page of the Naperville Smart Meter Awareness website features an Infowars segment where one of the ladies who was arrested for attempting to prevent the installation of the smart meter on her home is interviewed.

There are a lot of statements made in this video that I take issue with, but I'll keep this post focused on smart meters. This Infowars video segment references this article from Watchdog News Daily titled "Health Hazards Linked To Utility Meters". Personally, I'm dubious of any news site with pop-under advertisements, but I'll let it slide this time for sake of research. 

Joe Esposito from Owasso, Okla., had a smart meter installed on his home in 2011 as part of a pilot program developed by the Public Service Company of Oklahoma. Even though he asked that a meter not be installed on his home, Esposito found one mounted on the side of his house when he came home from work. 

It was then his health problems started. Esposito started experiencing dental problems, from aching teeth to a constant tingling sensation. He also started to experience aches in his leg which only got worse at night. 

After watching a video titled “Smart Meters & EMR: The Health Crisis of Our Time” by Dr. Dietrich Klinghardt, Esposito followed the advice in the video and installed some lead sheeting around the meter on the outside of his house. The results were dramatic. He had the first good night’s sleep in months and the pain in his leg was gone. Additional protection inside the home added later gave relief from many of his other symptoms. As an experiment, he would sometimes sleep without the protection and his pains would return.

This article features a correlation/causation between smart meters and health effects which references a video by Dr. Dietrich Klinghardt, but there is no link to the video in the article. I found a link to the video on StopSmartMeters.org.uk under the title "Dr. Dietrich Klinghardt – Smart Meters & EMR: The Health Crisis Of Our Time"

Dr. Klinghard makes the case that the increase in chronic medical conditions is related to the increasing electromagnetic radiation exposure (at the 6 minute mark in the video), he then takes his correlation to using the estimated number of wireless subscribers and comparing those numbers to the rise in Autism statistics. Correlation does not indicate causation, and I caution everyone to dig a little deeper when absolute statements are made with little regard for documenting information sources. It would appear that Dr. Klinghard has a vested interest in finding EMF exposure as the cause for multiple health symptoms as his personal website sells products that claim to reduce your exposure to EMF radiation.

Alice / Jan Svankmajer

At this point I'm stopping my research into the FUD that is out there about smart meters. I could go down this rabbit hole for untold iterations and still get back to the same conclusion I made several paragraphs ago.

The data I gathered clearly indicates that the RF output of smart meters (at least the L&G Focus-2 meter) is well below the FCC limitations and does not use 100% of the duty cycle when the smart meter is transmitting. I saw nothing in the spectrum analysis capture I performed that caused me to be concerned about the RF energy being transmitted by the smart meter installed at my residence. If you're wondering what your smart meter is or isn't doing, you can see it for yourself with a Metageek Wi-Spy 900x.




P.S.     I wish that Dr. Klinghard would say how much data I could store on his Computer Harmonizer K. Dwell Stick. For just a few dollars more, I can get a Kingston 128GB USB 3.0 Data Traveler on Amazon. The Kingston might even work as a noise filter or a protective pendant, but I'm sure of one thing it'll do - store a lot of data.

Meraki: Little Table - Big Ideas For Pokes, Fetches and Queries.

Sean Rhea of Meraki

Sanjit Biswas opened for the Meraki portion of Wireless Field Day 4. He told us that Meraki is still hiring and has started construction on a new office. At the time of the recording it had been only 45 days since the acquisition. Meraki is continuing to develop their product line, there are a variety of devices in the pipeline. The hardware development cycle is very long so it will be some time before there are Meraki/Cisco created APs. As a result of the acquisition, Meraki is now able to go International to Asia Pacific and other parts of Europe where they did not have a presence (previously Meraki was only in the UK and Ireland markets). Meraki is investing in adding richer features and their System Manager has been installed on about one million devices so far.


Cisco Update on the Meraki Acquisition with Sanjit Biswas from Stephen Foskett on Vimeo.

Sean Rhea took us behind the covers on Meraki's Dashboard. He has a PhD in distributed systems from UC Berkley and joined Meraki in 11/07.

Meraki (now part of Cisco) Cloud Architecture Deep Dive with Sean Rhea from Stephen Foskett on Vimeo.

Meraki customers partitioned across different servers, on different shards. This server is effectively a 1u raided server plus 1u backup (in a completely different data center/provider). One of the shards acts is the master and acts as a demultiplexer.

Thousands of Meraki devices connect to a shard.
100s of 1000s devices checking in per day
300GB of stats dating back over a year
New data is gathered from devices every 45 seconds.
If you are actively viewing the device in Dashboard it's every 1 second.

Engineering challenges are how to get access to devices behind NATs. The Meraki system mtunnel allows them to talk to their devices even if they're behind a NAT. The tunnel is fully encrypted and the information sent over the tunnel is SSL. Mtunnel is used across the entire Meraki hardware platform.
This custom secure tunnel requires only 2 packets/device/25 sec
It looks like ipsec, but is not point to point may route from a shard to a shard
Devices talk ssl over the tunnel. Each ap ships with a certificate with the shared secret on it.

The backend infrastructure is the same across switches and systems manager platforms.

Another problem is how to gather this information and minimize network overhead for Meraki & the customers and minimize the cpu costs for Meraki.

Meraki has developed a custom database tuned for statistical data. They achieve SSD-like speeds from inexpensive spinning hard disks.

Poder is a relatively new system at Meraki, first developed in 2009, then fully revampd in 2011. The idea is to gather stats from hundreds of thousands of client devices via the devices' internal web server.

The naive approach
• Overview
• Each device runs a small web server
• each shard runs one grabber daemon per statistic type (usage syslogs uptime etc)
• grabber fetches stats from devices as XML over HTTP.

Implementation
• 1 process per grabber, N+2 threads per process
• 1 thread to query DB for new devices to fetch from
• N threads to perform blocking HTTP fetches from devices
• 1 thread to insert fetch results into DB

Naive approach pros and cons
• TCP easily carries arbitrarily large responses
• Becomes expensive at scale
• A single HTTP fetch to empty webpage = 10 packets and 510 bytes
• Every grabber does its own fetch, no sharing connections
• Lots of threads for I/O parallelism = lots of context switches
• Limited to 1 fetch/node every 10 minutes.

A high-performance approach (4th gen)
• The first thing they created was an event driven RPC engine
• Non blocking IO, talks to 10k devices in 20 sec
• Uses UDP and Google protocol buffers (binary encoding format) (greatly reduces packet size)
• Stats are obtained through different modules
lldp modules (each has it's own thread)
• Database is read from by each of the different grabbers
Example: probing clients module/lldp module

A request object is passed through each of the modules, info requests are appended to request sent to the AP. Requests sent to APs often look different than switches due to the different functions that are done by different hardware. The modules are single-threaded, block on DB access and most are 200-400 lines of straight-line Scala code. There are no locking or synchronization issues to worry about.

Poder aggregates requests and responses for different modules. Overhead is a minimum of 2 packets/48 bytes for an empty response (80-90% less than previous Poder versions). Now Poder can fetch data every 45 seconds.

What to do with the data?
Where to store it and how to get it back out to display?
The storage cost of all the data that can be collected can be averaged out at 3600 clients (10 apps) (3 records per hour) 72bytes per record = 63gb per year.

Your standard OLTP databases (Postgres, MySQL)are not very good at clustering. Unless your working set fits in RAM queries take many seeks. A standard seek can take up to 18 minutes on a traditional hard drive to draw the overview dashboard graph.

With the custom database they've created (Little Table) it takes 2 seconds to draw (100mbs a second). Data can't be aggregated because aggregation discards useful information. The peaks can be averaged into data and no longer shows as severe/peaks.

Queries are written in order, writes footers with network_id and macs and are flushed every 10 minutes for durability (power outages) APs have statistics stored locally and can be queried again when the system comes back up. The files are merged and stored back to disk (in order) after 28 days (even multiple of seven to make the merge of data 'nice'). Max file size is 2GB to avoid messing with 64bit file pointers.

The Meraki data when it is on a Meraki server, it is unencrypted for speed. The data centers where their data is kept meets SAS 70 requirements. When the data is stored offsite for tertiary redundancy, the data is encrypted.

In my opinion, the Meraki presentation about their fine tuned database was the most technically interesting video of all of the presentations we watched during WFD4. I have very limited knowledge of database structures or data storage methods, but Sean held my attention and I think I almost understood what he was describing. Watch the video and see what you think.

Lomography Smartphone Film Scanner [unboxing and first scans]

I was an early backer of the Lomography Smartphone Film Scanner Kickstarter project. The scanner arrived a few weeks ago, but I'd been waiting to open it until the Lomography Scanner app was available via the iPhone app store. Well, the app was finally released yesterday, so today I opened it all up and gave it a whirl.

The film scanner is a little plastic gizmo that has a backlit lightbox at the base of the scanner (two AA batteries not included), several plastic expansion stackers (in order to adjust the closeness/focus of your smartphone camera) and an adjustable holder for your smartphone at the top. It also comes with a little foam shim to keep your phone level against the opening to the lightbox.

Your first choice is to select regular, lomokino or panoramic for your negative type. Then the scanner app gives you several options for the images you're scanning in: NONE, NEG, B&W, SLIDE, XPRO and REDS. It's not entirely clear what the negative settings are for and there isn't any context help built into the first release of the scanner application. I chose B&W for my negatives, but perhaps I should've chosen NEG. It's hard to discern the difference between the two settings.

There aren't any separate controls for white balance or exposure. I was able to adjust the white balance by touching opposite corners of the image before taking the photo, but it was hit or miss at best.

I couldn't find my shoebox of negatives, so I used some black and white negatives from my community college photo class (circa 1992). Making a print of every image I shot wasn't feasible at the time. Firstly due to the cost of the photo paper and secondly, not every image on the roll was good enough to print. Below is a slideshow of a few selected images I scanned in. There's one of Salt 'n' Pepa from the 1992 Inauguration festivities, a couple of self portraits, a sequence taken for a photo class project called "Portrait of Self Without Self", and a few of my first car (a 1964 Plymouth Barracuda).

The scanner device is pretty cool, even if it the smartphone app could use a little help. Now if only I could remember where I stashed that shoebox of negatives...

Geek Day is next week in SF! (March 27th - 28th)

Anyone in San Francisco next week?

I'll be at the World Wide Technology's (WWT) Geek Day event next week, deep in the heart of San Francisco!

Not registered yet? It's FREE!

Go get yourself a badge and come by and pick my brain about all things wireless!

"Geek Day is a technology tradeshow featuring live, working demonstrations from the industry's best-of-breed collaboration, wireless, mobility, security, datacenter and virtualization solution providers. 

The two-day agenda features the Geek Lab, where solution providers conduct on-demand demonstrations, and an agenda of breakout sessions where WWT and the Geek Day Sponsors provide deep dive product and solution information sessions."

Hyatt Regency San Francisco

5 Embarcadero Center

San Francisco, CA 94111

Cisco: 802.11ac Client Cards, Location Analytics and Snapdragons! #WFD4

Brian Hart (Strategic Initiatives Group) and Mark Denny gave us the 802.11ac update. We had a show and tell with the 802.11ac module (looks just like the monitor module. We also were shown two 802.11ac clients. A Linksys 1x1 802.11ac USB wireless adapter and a Netgear USB adapter.


Cisco Update on 802.11ac with Mark Denny and Brian Hart from Stephen Foskett on Vimeo.

The adoption timeline for CY2012 is consumer devices will be available from Linksys/Netgear in Q2. The first 802.11ac laptops will start shipping in Q4 2012. For CY2013, Wave 1 Q2 and 802.11ac mobile devices Q3 For CY2014, Wave 2 Q1, Client proliferation in Q2 and 802.11ac wave 2 starts to roll out in Q4 of 2014.

Multi User MIMO and the 4th spatial stream will be the big news items as 802.11ac progresses.

Jeevan Patil presented an overview on the 3850 Unified Access switch and the 5760 controller. The breakdown of the Cisco tagline "One Policy, One Management, One Network" is as follows:


Cisco Converged Access & Wireless Controller 5760 with Jeevan Patil from Stephen Foskett on Vimeo.

One policy = ISE
One Management = Prime
One Network = 5760 & 3850

These controllers are not replacing all the other controllers in the portfolio, these platforms allow for more options in designing a robust wireless infrastructure. The 3850 can support 1- 50 aps per switch/stack for directly connected APs, 2k clients per stack and a 40Gbps aggregate uplink capacity per switch. Converged access mode is where the access points terminate the data plane on the switch. This creates a single point of policy enforcement. The encrypted wireless user data can be acted upon as the data reaches the switch port instead of needing the encrypted data to get back to a centralized controller before you can apply QoS or utilize Application Visibility Control on the wireless data.

Controller licensing is moving to a Right TO Use (RTU) trust based model and RTU licensing is built into the universal images. There is a single CLI to turn on any license level, the customer agrees to the EULA and you can move licenses between 3850s and 3850s /and/ 3850s to 5760s /and/ 5760s to 5760s.

The connected mobile experiences was covered by Jagdish Girimaji and Mir Alami


Cisco Connected Mobile Experiences with Jagdish Girimaji from Stephen Foskett on Vimeo.

The goal of the connected mobile experience is to detect presence, connect customers and engage them. With the MSE & 7.4 code, retailers can get visibility into where the customers are within their stores and venues. Cisco recently purchased a company called ThinkSmart Technologies and is beginning to integrate location analytics into the MSE. It is possible to track non-associated clients if their Wi-Fi is turned on, yes - the client does not need to be associated to be detected and tracked. If a customer has downloaded a retailers application and have it installed on their smart phone, the Cisco infrastructure can automatically associate the client to the retailer's guest wireless network without the end user interacting with the app at all. The customer example is for Target stores, and the app can be made to automatically launch when the smartphone is detected as having the app installed and being in the proximity of the wireless network. Cisco has done a considerable amount of work with Qualcomm to embed the Mobility Services Application Protocol (MSAP) into the firmware at the silicon level (mostly Android devices). The new Android devices with the Snapdragon silicon in them will be able to take advantage of automatically discovering services (pre-association).

Pinpointing user locations is not currently possible, so stores will be divided into multiple zones. The accuracy they spec is 5m accuracy. In the next 6-9 months there may be improvements in that level of accuracy. The notification to the end user is done over the 3G or 4G prior to the user being associated to the 802.11 wireless network.

This level of tracking capability spurned a lively conversation about the security ramifications of having your wireless auto connect without you interacting with the application at all. The authorization to associate will probably be buried in the EULA of the app when you download it. Some users may take issue with their MAC address being detected and targeted for data pushes to the device. The Seattle Airport and Copenhagen airports have taken part in a study to determine how many devices pass through the airport at any given time. 

Mir Alami went over the infrastructure features of the Location Services Topology and explained how the users associate/authenticate and are interacted with via the MSE/custom built app. He walked through a demonstration of how the application would work and possibly send coupons/maps or other information to the targeted customer. The custom application does not have to be running and you do not need to be logged into the guest wifi to get this information push.

Aruba: VisualRF, AppRF and AirWave (Getting better all the time) #WFD4

Keerti Melkote presented the history of Aruba and told us they celebrated their 11 year birthday on February 14th! Aruba sees opportunity for innovation around client location awareness and utilizing software defined networking architectures in a campus environment. IPV6 will continue to be more and more important and will impact network designs in the future.

Aruba AirWave: Monitoring the Health of a Wi-Fi Network from Stephen Foskett on Vimeo.

Rob Gin (Aruba's AirWave expert) and Sujatha Mandava (Product Manager for AirWave) gave us a login to AirWave and we spent some time digging around viewing our client statistics and network utilization information. I made a few screen captures of views I found interesting. I could focus on a single client to see the information on client link, can focus/unfocus the display to show the access point information for just that single client device. The help desk view in AirWave does not allow you to adjust the thresholds per device types for alerts, but the Admin view will let you make those adjustments. Airwave tracks upstream devices and can determine which switch a controller is connected to in order to view wired/wireless data. They use the bridge forwarding table or CDP to gather information from switches. They can take anything in MIB2 to correlate information on the upstream device. The RF Performance views can start from the client perspective. The charts begin with information on clients with low SNR values. RF attenuation will be recalculated based upon the access points data that they can detect from one another. The colored lines to the clients from the access point shows the frequency (2.4 or 5GHz) that the clients that are connected on. The "Simulate failure" button shows the RF coverage without that single access point. You will be able to export client session reports to CSV and get them emailed to you in version 7.7 code.

VisualRF does location calculations on its own, unlike Cisco's MSE which is used to perform location tracking calculations. AirWave/Visual RF can take location information from Prime and use that information to place clients on the floor plan. Autoprovisioning of access points can be done for a defined region and calculated based on coverage needs (voice/data rates/signal strength). AirWave can be configured for specific triggers to alert on given events: Hard Drive space, RF utilization etc. The administrator can drag and drop the rules (from the Rules page in Airwave) to prioritize them on the fly (like mobile ACLs). Access point OUI information can be used to filter out rogues by a single vendor (like 2WIRE SSIDS). Airwave is capable of storing data for up to 5 years. Airwave is priced on device count. Access points, switches and controllers count as one device count.

AppRF in version 7.7 will provide information to AirWave (similar to Cisco's AVC) will show the top 3 destinations of client traffic, top trends for top 3 applications and uses NetFlow-like information.

Aruba Controllerless Wi-Fi from Stephen Foskett on Vimeo.

One of the access points at a remote site would have an https connection to AirWave for reporting on RF statistics. The number of possible users per controllerless group depends on vlan sizing. Aruba no longer recommend lots of access points in the branch managed by a controller in the data center and for the home offices, Aruba is still recommend having a controller at HQ managing the remote home users.

Aruba makes use of software managed AP purposing, instant APs, RAPs etc. The access point function is determined by the software that has been loaded onto the access point.

Ozer presented the evolution of the controller/controllerless architectures. There are many reasons why you would choose a controller based solution vs a controllerless solution. 

Aruba Controller-based and Controllerless Wi-Fi from Stephen Foskett on Vimeo.

Questions to ask your wireless vendor about their architecture options:

Can your controllers perform:

• Centralized encryption and policy enforcement?

• Local and centralized switching at the same time?

Can your controllerless APs:

• Self configure from the cloud

• Work without extra management software

Can i move from controllerless to controllers?

• With the same APs?

• Without going to the ceiling?

Can I mix and match architectures?

Scott Calzia (Product Manager of Aruba's Campus Controller product line) reviewed the features/functions of the 7200 series controller. It's the 3rd generation controller platform. There are 3 models of controllers: 7210, 7220, 7240. Each has four 1/10GB interfaces. The pair of dual media ports, can be used for interface connectivity, OOB or HA. Each has hot swappable load sharing redundant power supplies and field replaceable fan trays. There is an optional expansion slot *currently not in use*.

The highest end controller can support the following:

2000,000 firewall sessions

2048 APs, 32k devices 40GB

8 cores cpu, four cpus each.

SSD 8GB SD RAM 8GB EOS Flash memory

The controller hardware available now scales to support 4x the number of access points than it did previously.

The controller can support up to 40GB of encrypted throughput.

Balajee Krishnamurthy (Aruba TME) described AppRF as able to define policy decisions based upon applications detected on the wireless network. AppRF can identify applications based on ports and urls being used/accessed. deep packet inspection is possible and there are heuristics for lync, bittorrent, skype. AppRF can monitors the call setup and sync to differentiate Lync voice from Lync data (XML API). Lync voice/video over the air is prioritized, reporting in the Firewall dashboard doesn't have the differentiation to show the different data streams in Lync.

David Munro and Neil Kulkarni covered the Aruba Instant / controllerless solution.

The activate as a service is free for Instant AP deployments via activate.arubanetworks.com. If you have a virtual controller at remote location, additional instant aps discover virtual controller and download image and config from airwave management mode at the data center