Internet Network Security Blog

Ruckus may not be one of the biggest companies in the WLAN market, but it is certainly among the most innovative. The Ruckus antenna story makes those of us who care about things like coverage patterns and beam forming get all tingly, and its latest client-facing feature set aims to balance extremely high performance with minimal client pain for getting securely connected in interesting new ways. There’s a lot to talk about here.

The big story with Ruckus at the wireless network edge is a technology dubbed BeamFlex. Nowadays, every WLAN player that has a prayer of claiming market share has a feature called "beam forming." Simply put, beam forming varies the energy at the transmit antennas in ways that are meant to optimize and shape output signals for the benefit of each wireless client. There’s a lot of engineering-speak behind beam forming, but for our purposes, some methodology controls the manipulation of phasing out multiple antennas, and if it’s done properly, clients will benefit from better-quality signal and higher data rates. Despite marketing hype, there are lab studies that WLAN makers don’t always get it quite right.

In reality, there are different ways to achieve beam forming. Some are simple to understand and have debatable results when a wireless environment is populated with mobile devices that are used at every angle (and antenna orientation) imaginable. Then there’s Ruckus’ BeamFlex. BeamFlex uses adaptive antenna technology to take the notion of beam forming to a whole other place aimed at putting the competition to shame. Ruckus’ antenna magic is like no other on the market, of that there is no doubt. Whether you buy into the Ruckus product line or not, the company does a wonderful job of telling its own story through a number of videos and whitepapers that are refreshingly devoid of the typical marketing noise that pervades the enterprise wireless world. I recommend the Ruckus website at www.ruckuswireless.com, if just for the tutorials on beam forming and other highly technical wireless topics distilled down to very understandable form.

Moving downstream from the antenna, Ruckus has just announced the new FlexConnect feature set that simplifies provisioning secure access for the growing multitude of mobile devices likely to hit any corporate wireless network. Though WPA2 and 802.1x have become staples of the laptop-centric WLAN, getting a variety of device types configured at the supplicant level can be onerous and time-consuming. FlexConnect answers the challenge for Ruckus environments in a couple of different ways.

One of the FlexConnect tools called Zero IT is a framework for getting clients configured with proper supplicant settings (think Cloudpth’s XPressConnect functionality for loose analogy). This is where Extensible Authentication Protocol (EAP) type, credentials, servers and certificates, and certain Active Directory parameters, are set in what otherwise can take several steps to accomplish manually.

Also part of FlexConnect, Dynamic PSK creates per-user encryption keys that provide an alternative to RADIUS-based WPA encryption methodologies. (We’re seeing this offered by more Wi-Fi vendors as a convenient answer to clunky traditional client security methods.) There are numerous options to using per-user pre-shares, and when done properly they are great for self-service while meeting enterprise security requirements.

A tool called SpeedFlex also makes troubleshooting smart devices easier by running a battery of tests from each client when trouble is afoot. All in all, current Ruckus customers that have ZoneDirector controller-based networks will do well by adding the functionality of FlexConnect.

For those unfamiliar with Ruckus, the company is starting to make noise in several customer spaces, and is one to watch as a frequent innovator.

At the time of publication, Ruckus is not a client of and has no business relationship with Lee Badman.

Enterprises have made great strides improving the energy efficiency of processors, servers and storage technology, but an expert on eco-friendly IT says that attention also has to be paid to networking equipment and the addition of power management features of software. Lori Wigle, president of the IT industry-led Climate Savers Computing Initiative, spoke at a CIO conference on data center efficiency and cloud computing held last week near San Francisco.

The Initiative was co-founded by Intel and Google and includes tech industry giants such as Microsoft, VMware and Cisco Systems, as well as Dell and Samsung, which were co-sponsors of the conference. The group’s goal is to reduce carbon dioxide emissions attributed to data center operations by 54 million tons a year. So far, the four year old organization has reduced them by 32 million tons, said Wigle, who is also general manager of eco-technology at Intel.

The Initiative’s mission is to address power delivery to data centers and introduce power management to software, she said. The group pushes vendors to make more energy efficient products and gets buyers to pledge to purchase IT based on its energy efficiency.

“We haven’t done a good enough job of working with the application vendors, whether they’re the big ISVs [independent software vendors] or even the software that you develop internally for your businesses,” she said. “We’ve started reaching out to software developers as another audience to educate them on what it means to have energy smart software.”

While significant improvements in server energy efficiency have been realized, primarily through the adoption of virtualization and the development of energy efficient processors, networking equipment needs to improve, too, Wigle said. Networking equipment “hasn’t had as much focus historically,” she said, but added that a group within the Initiative, including Cisco and Juniper Networks, is focusing on that issue.

“These are arch-rivals working together,” she noted. HP, which increased its competitive pressure on market leader Cisco in networking, is not in the network equipment group, but is a member of the Initiative, she said in an interview.

Energy efficiency is also important for reducing the waste in IT from electricity that comes into the data center but doesn’t actually power any computing. Electrical efficiency is measured by a Power Usage Efficiency (PUE) rating. The ideal is to achieve a rating of 1.0, which means each watt of electricity equals 1 watt of computing. The higher the PUE, the less efficient a server, desktop computer, or electrical converter is.

Bill Weihl, Google’s energy czar, said at the conference that by improving energy efficiency at five of its data centers, the company reduced their PUE rating to 1.5 from 2.4. Samsung touted the energy efficiency improvements in its dynamic random access memory (DRAM) and solid state disk (SSD) storage products.

An older model DDR2 memory device consumed 102 watts of power, but a new DDR3 introduced in 2010, consumed only 14 watts, an 86 percent reduction, said Jim Elliott, vice president of marketing at Samsung. Also, Samsung sold its hard disk drive (HDD) business to Seagate Technology earlier this year to focus on SSDs, which while more expensive than HDDs, are significantly more energy efficient, said Elliott.

SSD’s, based on Flash memory technology, use 75 percent fewer watts than an HDD in active mode and 87 percent fewer watts when the disks are in idle mode. “No moving parts means lower power and heat,” he noted.

See more on this topic by subscribing to Network Computing Pro Reports Parallel Memory Optimized (subscription required).

Enterprises have made great strides in improving the energy efficiency of processors, servers and storage technology, but an expert on eco-friendly IT says that attention also has to be paid to networking equipment and the addition of power management features to software. Lori Wigle, president of the IT industry-led Climate Savers Computing Initiative, spoke at a CIO conference on data center efficiency and cloud computing held last week near San Francisco.

The initiative was co-founded by Intel and Google, and includes tech industry giants such as Microsoft, VMware and Cisco Systems, as well as Dell and Samsung, which were co-sponsors of the conference. The group's goal is to reduce carbon dioxide emissions attributed to data center operations by 54 million tons a year. So far, the 4-year-old organization has reduced them by 32 million tons, says Wigle, who is also general manager of eco-technology at Intel.

The initiative’s mission is to address power delivery to data centers and introduce power management to software, she says. The group pushes vendors to make more energy-efficient products and gets buyers to pledge to purchase IT based on its energy efficiency.

"We haven’t done a good enough job of working with the application vendors, whether they're the big ISVs [independent software vendors] or even the software that you develop internally for your businesses," she says. "We’ve started reaching out to software developers as another audience to educate them on what it means to have energy-smart software."

While significant improvements in server energy efficiency have been realized, primarily through the adoption of virtualization and the development of energy-efficient processors, networking equipment needs to improve, too, Wigle says. Networking equipment "hasn’t had as much focus historically," she says, adding that a group within the initiative, including Cisco and Juniper Networks, is focusing on that issue.

"These are arch-rivals working together," she notes. HP, which increased its competitive pressure on market leader Cisco in networking, is not in the network equipment group, but is a member of the initiative, she says.

Energy efficiency is also important for reducing the waste in IT from electricity that comes into the data center but doesn’t actually power any computing. Electrical efficiency is measured by a Power Usage Efficiency (PUE) rating. The ideal is to achieve a rating of 1.0, which means each watt of electricity equals 1 watt of computing. The higher the PUE, the less efficient a server, desktop computer or electrical converter is.

Bill Weihl, Google’s energy czar, said at the conference that by improving energy efficiency at five of its data centers, the company reduced its PUE rating to 1.5 from 2.4. Samsung touts the energy-efficiency improvements in its dynamic RAM (DRAM) and solid state disk (SSD) storage products.

An older-model DDR2 memory device consumed 102 watts of power, but a new DDR3 introduced in 2010 consumes only 14 watts, an 86% reduction, says Jim Elliott, VP of marketing at Samsung. Also, Samsung sold its hard disk drive (HDD) business to Seagate Technology earlier this year to focus on SSDs, which, while more expensive than HDDs, are significantly more energy efficient, says Elliott.

SSDs, based on flash memory technology, use 75% fewer watts than an HDD in active mode and 87% fewer watts when the disks are in idle mode. "No moving parts means lower power and heat," he notes.

See more on this topic by subscribing to Network Computing Pro Reports Parallel Memory Optimized (subscription required).

Application delivery network provider F5 Networks has acquired intellectual property (IP) and certain technology assets from Crescendo Networks. Chief among the assets are a number of Crescendo Networks engineers at its headquarters in Tel Aviv, Israel, who will now join F5 engineers already based in that city.

A news release quoted an F5 executive as saying that Crescendo’s technology provides layer 7 field-programmable gate array (FPGA) capabilities for hardware and security solutions. With the acquisition, F5 said it will be able to further strengthen its products’ ability to address the exponential growth of Internet traffic and the rising number of security threats to software applications.

“Layer 7 is at the application layer,” explained Cindy Borovick, research vice president for data center networks at IDC. “What F5 found in Crescendo was that there was some particular IP related to accelerating different types of traffic in hardware and about where you could offload from the [central processing unit] CPU and do some traffic processing in the hardware.”

The FPGA is an integrated logic chip that can be programmed by the operator. Once the design is set, these hardware chips can be produced in bulk for faster performance. While Crescendo has won “a number of industry awards for its innovative products,” F5 said, the company was unable to make a go of it in the market.

F5 said the acquisition was made “through a liquidation process in Israel” of Crescendo, which was founded in 2002. An F5 spokesperson explained by e-mail that liquidation in Israel is a court-ordered dissolution of an insolvent company. It is not a “liquidation” as the term is used in U.S. Bankruptcy Court because in Israel only individuals, not companies, can declare bankruptcy. Terms of the acquisition were not disclosed.

“They were never really able to capitalize on the market opportunity,” said IDC’s Borovick of Crescendo. “They had a number of fits and starts in terms of marketing and they were never able to capture the market opportunity in North America.” The market has also consolidated with F5 as the largest supplier, Borovick added.

F5 Networks was placed in the vaunted upper right quadrant of “market leaders” in a November 2010 report ranking application delivery controller vendors from the research firm Gartner. Gartner put F5’s market share at 47 percent, highest of all the vendors.

Also ranked highly with F5 as market leaders were Citrix Systems and Radware. Crescendo was placed in the lower left quadrant of niche players and smaller firms. The Gartner study noted that while Crescendo had built an account base of Web 2.0 sites and other high-traffic Web sites, “Crescendo has achieved limited market and brand visibility in the enterprise.”

See more on this topic by subscribing to Network Computing Pro Reports Strategy: Bringing APM to the Cloud (subscription required).

VMware-based virtualization could get a big boost with Astute Network's ViSX G3 accelerator, which provided one early adopter with 1,500 percent improved read performance on critical applications, accelerated performance on its most demanding VMs and delivered a noticeable increase in the responsiveness of its database applications. The customer, Visioneer, also reported that backups were completed faster, with less impact on other operations, and productivity and efficiency are up as well.

Company executives says this is a new category of solution, network performance flash, which makes virtualization affordable for smaller businesses. It allows customers to dial in the level of performance they need for specific applications and guarantees them the level of performance they're looking for.

Available in three models, starting at $29,000, the VMware-certified ViSX G3 is a combination of enterprise flash memory, the company's DataPump Engine processor technology, custom hardware and software. Astute says the DataPump Engine processor delivers fully offloaded and accelerated network traffic (TCP/IP) and virtualized data store traffic (iSCSI), sustained flash performance, and multi-level RAID protection, the sustained performance equivalent of 100s of enterprise disks.

Mike Karp, VP and principal analyst, Ptak Noel Associates, says this solution has the potential to be a great enabler for virtualization, whether that virtualization is inside the data center or resides in the cloud. “Its key strengths lie in several areas. First, it's the easiest way I know to improve performance in a virtualized environment. The process is essentially this: plug it into the Ethernet, and walk away. Second, it's relatively agnostic, and out-of-the-box works with whatever network storage is available. Third, it provides increased IOPS performance across the network – not just to a single machine – and it does this at an aggressive price point.”

Storage and virtualization solution providers are keenly aware of the IO performance issues impacting virtualization implementations and are actively developing solutions to off-set these challenges, but they’ve a ways to go yet, says Steve Brasen, managing research director, systems management, Enterprise Management Associates. “The new ViSX G3 platform could resolve this problem today and aid organizations struggling with 'VM stall', boosting virtualization success rates and increasing the rate of adoption.”

Karp's major concern is that this can be considered an 'offload engine', which have typically had a bumpy ride in the marketplace over the last 10 years. “Historically they have been expensive add-ons that have been added one-by-one to individual workstations or servers, and while they could scale, it was expensive to do so because you had to add new a one to each node where you wanted improved performance. Astute Networks will have to make sure people understand that while ViSX shares a common goal with the older offload engines, this is fundamentally different from that sort of technology: this goes out on the network, and can be shared/leveraged across whatever processes are being transacted there.”

Brasen likes the fact that as an appliance-based platform, the ViSX G3 is a turn-key solution that is completely self-contained and can be easily and quickly deployed in new environments. “Although the initial release of the appliance is designed to support VMware implementations, future editions will provide more heterogeneous support, providing a single solution for supporting multiple virtualization deployment types. This heterogeneous support is critical as the majority of virtualization adapters (79% according to EMA primary research) utilize more than one virtualization platform.”

His caveat is that as an independent architecture, it adds an additional infrastructure component to the servers, storage, and virtualization software platform that needs to be purchased and managed. “As is common in any virtualization implementation, organizations need to choose between performance and complexity. Though, in this case, I would argue that the performance boost far outweighs the increase management impacts.”

See more on this topic by subscribing to Network Computing Pro Reports The Data Mastery Imperative (subscription required).

Looking to take advantage of what it calls a perfect storm, networking vendor Adtran, Inc. has bought privately held Bluesocket, which sells wireless network solutions with virtual control, for an undisclosed amount. This acquisition was effective August 4, and includes all intellectual properties, technologies and the established customer base, over 3,000 in 47 countries. The Bluesocket team and product portfolio will be incorporated into Adtran's Enterprise Networks Division.

The companies say there are three developments that are creating major discontinuities in the networking market: the explosion in 802.11n adoption (by 2015, it will account for over 90 percent of access points); the exponential increase in wireless demand (mobile devices sales were up 16 percent to 428.7 million units in Q2, according to Gartner); and the ability of cloud virtualization and WLAN delivery to obsolete controller-based implementations. Throw in sustainability, the ability to reduce power by 80 percent, and the only thing holding back the new entity is the ability to execute, notes Zeus Kerravala, senior vice president of research, Yankee Group, in a blog.

'Because of mobility, consumerization, and device evolution, we're clearly trending towards wireless as the primary network, which is creating another transition point in the industry.' He believes that market share change only happens at points of market transition, and this shift to wireless as the primary network creates a great opportunity for vendors in the wireless LAN space with alternative solutions.

'So now for Adtran to take advantage of this new asset, they'll need to step on the gas and be a lot more edgy and marketing-focused than they have in the past.' They are a very well run company, he says, but their marketing prowess, particularly on the enterprise side, is somewhat limited. As he doesn't expect the company to change their marketing roots overnight, he suggests a good interim step would be to work with the likes of Microsoft, Citrix and VMWare to build demonstrable case studies of how the Bluesocket solution provides a distinct benefit for things like VoIP, video and desktop virtualization. 'Bold move for Adtran, now it's time to go execute.'

A month ago Adtran reported sales up 23 percent year-over-year to $184,227,000 for the second quarter, while net income increased 33 percent to $36,943,000. By acquiring Bluesocket, it gets to tap into both the wireless and expected converged network markets, says Adtran's Gary Bolton, VP global marketing. He says they believe the wireless and wireline networks must be combined into one, and because moving to wireless means users will need to look at new solutions, this puts the new combined entity in an enviable position.According to the latest report from Dell'Oro Group, the WLAN market leaped 18 percent in the first quarter of 2011, with Aruba and HP demonstrating the fastest enterprise growth over last year. “While down on a sequential basis, the enterprise WLAN segment exhibited a growth rate of 22 percent year-over-year, reaching $573 million in revenues for the quarter,” stated Loren Shalinsky, senior analyst of wireless LAN research at Dell’Oro Group.

IDC says the global WLAN market came in just under $5 billion in 2010, growing 12.4 percent for all segments, with the enterprise segment leading the way up 34.9 percent year-over-year. It says the tremendous momentum behind smart mobile devices and the continued proliferation of higher performance 802.11n networks are driving enterprises to move forward with upgrades, extensions, and replacements of their wireless infrastructures.

As significant as the technology upgrade cycle underway for 802.11n, and the mobility explosion, it was the requirement to scale which drove Adtran to Bluesocket, says Bolton. “Cloud virtualization changes everything... and if you look at scalability as one of the most critical parts of the network... scale is paramount.”

Bolton agrees with Kerravala that execution will be key, and they don't have an infinite window to take advantage of this 'storm'. “We're hitting the ground running, and already holding integration road map meetings right now. We will be launching the first phase of our WLAN in the next few weeks.” Looking ahead, it's not just about seamless networking for both wired and wireless, but also about virtualized network management, he adds.

See more on this topic by subscribing to Network Computing Pro Reports Fundamentals: Wireless Mesh Networks (subscription required).

Hewlett-Packard invited reporters and industry analysts into its HP Labs research facility on Tuesday to share work it’s doing to make IT networks faster, more scalable and reliable, including replacing copper wire computer cables with optical interconnects. The open house at HP Labs comes as the technology giant goes after the network equipment business of industry leader Cisco Systems and as HP portrays itself as an innovator and not just a low-cost alternative to Cisco.

HP Labs researchers detailed work they are contributing to an industry standard project it calls OpenFlow networking, which aims to add a “control layer” to an IT network between the existing network management layer and the physical infrastructure of network switches, routers and other IT assets. The control layer would reprogram switches to identify certain data packets and take specific action based on that identity. A file transfer protocol (FTP) packet can be treated one way, while a hypertext transfer protocol (HTTP) packet can be treated another. The control layer intelligence would be able to reconfigure switches more quickly than a technician doing a tedious command line interface (CLI) reconfiguration, basically rewriting lines of code, said Charles Clark, a distinguished technologist in HP’s Networking business unit.

“In this system, what we’re able to do is more dynamically change the network through an OpenFlow interface in order to program the network,” Clark said, adding that OpenFlow capability can be introduced to an existing network infrastructure. “We can deploy new functionality in the management and control plane layer that causes those devices to do something different than they did before.”

HP is a member of the Open Network Foundation, a collaboration of multiple tech companies whose mission is to improve networking through software-defined solutions. Leaders of the foundation are Deutsche Telekom, Google, Facebook, Verizon, Microsoft and Yahoo. The networking industry is also represented; besides HP; Cisco, Brocade and Juniper are members.

Also at HP Labs, researchers are studying optical technology as a replacement for copper wires within networks to dramatically improve bandwidth and scalability, said Mike Tan, distinguished technologist in HP’s Intelligent Infrastructure Lab.

If an IT staff wants to increase network capacity to 25 gigabits per second (Gbps) from, say, 10 Gbps, they would have to reconfigure the electrical backplane in each traditional copper wire-based switch. “It needs to be reoptimized, redesigned and then retweaked,” Tan said.Optical technology eliminates the need to redesign the backplane and eliminates the need for the application-specific integrated circuit (ASIC) fabric layer for routing data streams among line cards within a switch.

To make that routing process quicker, the optical solution takes a different approach, he said. “What we have done is, in order to eliminate that, we’re broadcasting our traffic so that each line card broadcasts to all the line cards that are attached to the backplane,” he said. Each packet contains a header identifying the type of packet and how it should be routed.

HP is from two to five years away from having a commercial optical product, Tan said, and explained the lab is trying to take out some of the costs from developing the optical technology.

HP entered the networking market in earnest following its acquisition of 3Com, which was completed in 2010, and it has claimed some market share gains at the expense of Cisco. Cisco has countered by saying that HP’s competing largely on price, offering customers only a “good enough network” that won’t prepare customers for the growing demands being placed on enterprise networks.

But HP Labs is working on innovations that make networks not only more scalable but easier to manage, said Saar Gillai, vice president of the Advanced Technology Group and chief technology officer of HP Networking.

Referring to the complexity of the CLI process required to reconfigure a switch, Gillai said HP is innovating to make networks less complex. “It shouldn’t be this hard and part of the reason maybe it is this hard is that there hasn’t been enough competition to push the providers to provide better solutions,” he said. “Well, we’re changing that.”

See more on this topic by subscribing to Network Computing Pro Reports Fundamentals: The Switch to IPv6 (subscription required).

When management suites first emerged, companies were concerned with items such as, “Is the server up or down?” or “How much network bandwidth is being used?”. With those issues now sitting firmly in the rear-view mirror, businesses are looking for new types of management solutions. Datotel LLC, an IT data center services provider, based its recent purchase on an emerging area: “We needed to get a better understanding of our energy usage,” noted David Brown, President of Datotel, which now finds itself at the forefront in the use of such products.

In business since 2004, the company operates a 34,000 square foot data center in St Louis, MO that provides businesses with co-location services, managed services, and cloud computing resources to enterprises. The corporation, which has a staff of about 40 employees, started out using Dell Inc. servers to support its services and recently began relying on VMware Inc. virtualization software.

Datotel’s data center has 15T of storage and more than a 1,000 devices, whose deployment often occurred in a hodgepodge fashion. The corporation started with legacy hardware (mainly Wintel and UNIX systems) and added new solutions whenever it landed a new client.

The business managed the various data center elements in an autonomous manner, with a variety of tools examining the distinct elements. In 2007, the corporation wanted to get a more comprehensive view of what was happening in the data center, so it could proactively oversee its resources. The IT service provider looked at management solutions from CA, Hewlett-Packard Co. and IBM.

Datotel quickly realized that the power used by its data center systems, backup generators, and Uninterruptable Power Systems (UPSs) represented a significant expense, one rivaling its hardware and personnel expenditures. But the corporation did not know how much energy it consumed until the local power company’s bill arrived at the end of the month. In addition, the staff was not sure how the various IT data center products devoured electricity and had zero visibility into usage among energy related devices, such as racks, generators and cooling units.

Therefore, the IT services provider had no insight into how much energy its customers used. Consequently, the corporation established fixed monthly fees for different data center devices, so energy hogs paid the same rate as energy savers. Compounding the problem, customers frequently added servers, moved applications, and retooled their configurations, often without any input from Datotel personnel. As a result, some systems were not optimally configured and wasted energy.Addressing these issues would enable Datotel to deploy its staff more efficiently, reduce its expenses, and make its service more attractive to potential customers. To make that change, the company needed a more detailed understanding of how energy was being consumed. But with thousands of disparate devices strewn within the data center, collecting and collating usage data represented a considerable challenge.

After examining its options, the IT service provider selected a CA solution because its ecoMeter energy management system provides real-time and historical consumption data. The deployment began in early 2007 and took several months. Management systems are not plug-and-play. Instead, they require a great deal of customization, so businesses can collect and analyze relevant performance data. “A company needs to make a distinction between what is interesting to monitor versus what is useful information to collect,” noted Brown.

The change helped Datotel better manage its energy consumption and reduce its costs. Usage information is available in real time and is broken down by categories, such as device and time of day. The new system increased employee efficiency. Rather than have data center technicians tinkering with all of the devices to extract usage information, energy collection is now automated, so technicians work on other tasks, such as configuring new systems or troubleshooting problems.

Datotel shares the energy usage data with customers via an online portal. Since clients are now billed based on their energy usage, they are able to cut their expenses by implementing sound energy policies.

The IT services company is satisfied with the new system but would like one improvement. “More granularity is an ongoing request; we would like to be able to slice and dice more energy usage statistics, stated Brown.

Chances are such features will emerge in the future. Energy management is one of the new elements being woven into management suites. The tools are in an early stage of development but are expected to mature quickly as early users, like Datotel, become more familiar with them.

See more on this topic by subscribing to Network Computing Pro Reports The Data Center Balancing Act (subscription required).

Just back from a whirlwind two weeks working in both Great Britain and Haiti, and I feel compelled to reflect on the specifics and profound differences of creating WLAN environments in each location as I document the efforts. Certainly the cultures are dramatically different, but so were the solutions used, and neither is what I typically deal with in my day-to-day network administration duties. It’s been interesting, to say the least.

Let's start with London. My university (Syracuse) has a remote site in London, and for years it has limped along mostly as an island when it comes to IT. It turns out the site has a respectable wiring infrastructure, but its switching, wireless and overall networking approach was disjointed, undersized and underperforming, to the point where users would often leave the building and go elsewhere to actually get work done over the network. After a site visit from one of our managers, the right words were said and my group was on the hook to make things right for our faraway colleagues by creating a solution that allowed for us to monitor the environment from across the pond while giving the London folks some much-appreciated local administrative capabilities.

Though we are generally content with our Cisco 3,000-access-point, lightweight wireless environment on the main campus, our London center needed a complete network solution with a minimal number of boxes (and burden) required. We opted for Meraki's cloud-managed wireless and wired networking after doing an eval and comparing cost and complexity versus features and ease of use for different options. Though it's too early to say that Meraki was a slam-dunk for us in this case, all indications and feedback so far have me feeling quite good about how we proceeded.

On the ground, with only four days to complete what might typically be a couple weeks of work, I was able to meet the folks on the other end for the first time, verify and adjust the operational philosophy that would shape the technical aspects of what we were about to do, and wipe out a bunch of networking. I didn’t have to pull cable per se, but I did have to do a fair amount of rerouting of existing cabling as 35 MR16 access points, four Cisco network switches and two Meraki MX-70 boxes were brought to life and connected via site-to-site VPN back to the main campus. Training our man in London to do things like wireless guest management, static IPs for select devices, quarantining problem clients and traffic shaping was a snap through Meraki’s dashboard. Though I left London exhausted, I was quite pleased with the results.

Roughly 30 hours after catching a cab to Heathrow, I touched down in Port-au-Prince with three other full-timers from Syracuse University and seven students. This was my second trip to the impoverished nation, and we arrived with the goal of lighting up wireless networking on at least two of the University of Haiti’s campuses. The challenges would prove to be many.

We shipped three pallets full of network equipment, tools, snacks to munch during long days, and things like fasteners and tie wraps for the inevitable MacGyver-style work that would be needed. Our network solution would include Cisco a/g access points and switches removed from our campus during upgrades to 11n, and two brand-new BlueSecure controllers generously donated by Bluesocket that would provide the magic in the middle for simple local user authentication, rate limiting to preserve miniscule ISP links and a control point for UEH staff.

To jump to the end of the story, we were quite successful despite one obstacle after another. A couple of days were lost mostly to customs frustrations, a tropical storm threat slowed us down, and routine issues like flaky power and lack of a real business IT frame of reference by our hosts needed to be worked through. Our students had a ball taking part in mapping each campus, assisting with component configuration and doing installation work. Though the Port-au-Prince networks ended up being quite simple compared with the London initiative (basic WEP on the access points and local user authentication in the Bluesockets), the functionality we delivered at three UEH campuses was easy to train people on and light years ahead of what little was in place. As the UEH folks gain networking savvy, we’ll also be able to evolve the networks on existing hardware toward a more secure and robust framework. Baby steps.

Back home, it feels like I’ve been gone for a year. I’ve met some great people in both countries, had to really think fast in different ways in both places, and have thoroughly enjoyed bringing up WLANs we can be proud of in different parts of the world. Every now and then it’s nice to be able wander off the reservation.

For the first time in history, we have an economic recovery without corresponding rehiring because corporate automation and other factors are allowing companies to function leaner. It’s a frightening change for some, but it also opens opportunities for IT professionals who are gifted in analytics, workflow re-engineering, vendor management and integration. One case in point is warehousing and what has happened there to allow companies to respond rapidly to market conditions while keeping costs down with IT.

It begins with a sea change in how companies think about warehouses. In the 20th century, warehouses were perceived as a "necessary evil," packed with processes that were largely manual and inventory that frequently had to be expedited at the last minute because the warehouse wasn’t integrated with marketing and planning. ERP (enterprise resource planning) systems came on the scene to address this, but still the warehouse lagged.

Twenty-first century warehouse system thinking is different. Suddenly, warehousing is in nearly every company’s strategic goals. It is being viewed as an area of competitive advantage. Some of the pressures for change are due to increased global competition for customer wallet share, but equally important is the continuing growth of e-commerce, which thus far in the 21st century is seeing an annual growth rate of 12% to 14%. From an IT standpoint, this requires an integration of multiple shipping channels in real time with the warehouse.

But perhaps the largest warehouse sea change is philosophical. Companies are systematically moving their warehouses to the "front lines" of the customer experience, recognizing that instantaneous order fulfillment (and customer gratification) are crucial (and difficult to achieve) in e-commerce, and are only attainable with 21st century warehouse technologies. "Investing in your warehouse is not just about cost reduction anymore," says Mitch Rosenberg, VP of marketing at Kiva Systems, a supplier of automated material handling order fulfillment systems. "Twenty-first century consumers are different than their 20th century predecessors, so the warehouse must also be able to contribute to a beneficial customer experience."

Delivering the right goods on time to an expectant e-commerce customer often means that the warehouse distribution center becomes the "store," and the end customer experience becomes the experience that is happening within the warehouse. This transformation won’t happen overnight, as industry experts forecast that by the mid-21st century, brick-and-mortar retail transactions will still comprise 70% of all sales.

However, companies planning for the future are already funding and incorporating new warehouse technologies that improve efficiencies and responsiveness to market demand. "It’s all about customer gratification," says Rosenberg. "In the 20th century, companies worked on this by optimizing their stores so that customers could walk out of these stores with the items they had purchased. Now, people want this same kind of instant gratification by immediately obtaining their goods through e-commerce, but this is hard to get online. For e-commerce to compete with traditional brick and mortar retail in an area like instant customer order fulfillment, warehouse distribution centers must be optimized to deliver on that promise."

So what are the technologies and automation that can bring all of this to pass?

Batch-and-wait operations With this type of operation, orders are system-batched up during picking to improve process speed and then automatically separated into customer orders before leaving the warehouse for the customer. This reduces the latency factor that now exists when orders are separately picked for each customer, and it improves the overall performance of the distribution center in website-to-doorstep order fulfillment.Improved visibility of goods flows into and out of facilities This is achieved through integration of the warehouse with the systems of outside logistics partners. The integration allows companies to balance transportation and other costs with customer service levels.

Mobile robots and movable shelving units In this model, the shelves come to the workers instead of the workers to the shelves. This makes a big difference when you are talking about warehouses that are city blocks long, where workers walk 5 to 15 miles a day on concrete floors. In the automated system, robots the size of living room ottomans "walk" inventory over to the worker. The inventory identifies itself to the worker, and the worker picks and packs it. The end to end process is highly ergonomic, safer and faster.

New automation for forklifts and trucks "The forklift industry is an emerging field," says Joe LaFergola, manager of business and information solutions for Raymond Corp., which provides material handling solutions. "Over the years, people have tried to apply metrics to understand what’s going on in the warehouse. In the early days, they used analog measuring devices. Lately, the degree of forklift integration has produced vehicle management with onboard computers. This makes it easier to extract data from the forklift, which can be transported over the internal wireless infrastructure of the warehouse to a centralized system over an open API for ease of integration." The forklift vehicle manager sends a series of data codes that are actually hexadecimal data points. Each data point references a particular parameter in the truck, such as travel, lift, dead man hours, key hours and so on. The data flows though a warehouse gateway that is web-based, so all the warehouse manager needs to use it is a browser.

This combination of technologies and automation will redefine key performance indicators (KPIs) for warehouses, emphasizing labor elimination, automated and highly integrated systems, and accuracy. The business will be looking for:

Labor cost per order on shipment

If you can get four times the shipments out the door without having to hire more personnel, your labor costs are down 25%.

The cost of exceptions

If a conveyor belt breaks, what’s the mean time to repair (MTTR)? How much of your operation is affected? What’s the upstream and downstream impact?

Manned versus unmanned travel time on trucks

Manned travel time is $30 per hour. Unmanned travel time is $4 per hour. With robotics, 750 miles a day in manned travel time can be reduced to 250 miles a day.

All of this adds up to business driving technology adoption. Now, it is IT’s job to likewise transform its thinking into more of a process engineering approach that integrates people, facilities and automation with the traditional software, hardware and networks IT is accustomed to working with.

See more on this topic by subscribing to Network Computing Pro Reports Strategy: Automation and Orchestration (subscription required).