The Challenge of Wireless Data

Details
    Published on Tuesday, 14 August 2012 05:49

Since the U.S. government began assigning broadcast rights in the 1920s, the process has worked on a simple principle: A licensee received the exclusive right to the use of a slice of wireless spectrum, usually limited to a specific geographic area. Today, we are facing a big problem. All of the usable spectrum has been allocated, but the explosive growth of smartphones and other wireless devices is creating rapid growth in the demand for bandwidth. Technology that allows existing spectrum assignments to be used more efficiently has helped considerably. The government has also released some existing bandwidth for wireless data use, most recently a big chunk of 700 MHz spectrum that was freed by the transition to digital television.


But neither spectral efficiency nor the slow process of freeing and reassigning spectrum seem adequate to meet soaring demand. A recent report by the President's Council of Advisors on Science & Technology (PCAST) concluded that technology that allows spectrum to be shared among multiple radio systems will be needed: "Spectrum should be managed not by fragmenting it into ever more finely divided exclusive frequency assignments, but by specifying large frequency bands that can accommodate a wide variety of compatible uses."

The Federal Communications Commission, which shares responsibility for spectrum assignments with the Commerce Dept.'s National Telecommunications & Information Agency, is responding. At a recent forum, FCC Chairman Julius Genachowski said: "We are still operating on the same assumptions and parameters as when we began. But the world has changed."

The FCC is currently working on strategic "band plans" for two chunks of spectrum that are going to become available. One is for bandwidth that will be cleared as TV broadcasters agree to auction off spectrum they hold in the 600 MHz band. This will take a few years, because the process mandated by Congress is complex. But it is prime spectrum for wireless data and new technologies will increase the efficiency of its use.

Another band being considered, at 3.5 GHz, poses different challenges. In general, the higher the frequency, the shorter the range of radios using it, and current cellular-type radios operate at between 800 MHz and 2.1 GHz. One idea is to make a virtue of necessity and use the higher frequency spectrum to build large numbers of very small cells, greatly increasing the reuse of channels in any given area. (Wi-Fi and Bluetooth work in much this way.) But there is an additional challenge in the 3.5 GHz band. Certain types of military radar are currently using the band. So the radios will have to avoid interference with these vital installations by respecting "exclusion zones" that bar the use of specific frequencies in specific locations.


To do that, one possibility being debated is the use of "smart," "agile," or "cognitive" radios that are able to operate on a wide variety of frequencies, using software to sense unused and available channels. One recent twist on cognitive radios is to have them interact with a secure database that will warn them away from protected radio systems, such as military radar.  Still, many issues remain to be addressed. For example, one concern expressed by participants in the FCC forum was whether these new, more complex radios could come close to matching the battery life we have come to expect from mobile devices.

These are early experiments in what is likely to become a much broader move toward spectrum sharing as a way to use bandwidth more efficiently. And many critics have doubts about the reliance on unproven technologies. Writes Richard Bennett of the Information Technology and Innovation Foundation: "Advances in technology are often messy and disruptive, so it's much more sensible to continue pursuing a multi-faceted strategy that allows technologies to compete on the basis of the value they offer consumers than to tilt the scales in favor of one and only one system."

Wireless carriers, too, would rather the government just surrender some of its spectrum for wireless data use. Christopher Guttman-McCabe, vice president for regulatory affairs of CTIA: The Wireless Association says that while "it is sensible to investigate creative approaches for making federal government spectrum commercially available, including the development of certain sharing capabilities," the CTIA believes "the gold standard for deployment of ubiquitous mobile broadband networks remains cleared spectrum."

The problem is that it is growing increasingly hard to find spectrum that can be easily cleared. The time and effort needed to clear bandwidth means we will need major technological advances in efficiency to close the gap, including potentially the development of more ubiquitous frequency-sharing technologies.

By Steve Wildstrom

The contents or opinions in this feature are independent and do not necessarily represent the views of Cisco. They are offered in an effort to encourage continuing conversations on a broad range of innovative technology subjects. We welcome your comments and engagement.

Used with the permission of http://thenetwork.cisco.com/.

Read More

Mitigating BYOD Risks on Network

Details

Published on Tuesday, 17 July 2012 16:03

Bring-Your-Own-Device (BYOD) trend has resulted in a growing number of employee-owned devices on the network. Unauthorized use of the corporate network leads to security issues, as well as potential disruption of the network. In order to mitigate some of the risks on corporates networks, Blue Coat has introduced mobile application controls.


As employees increasingly bring their own mobile devices to work, the mobile app security gap has widened. The vendor says its new solution gives IT the ability to manage the use of unsanctioned applications and enforce policies across all devices on the network. The growth in BYOD initiatives has driven an increase in mobile devices and unmanaged applications on the corporate network.

According to Forrester Research, 57 percent of employees chose and purchased their own smartphone without any direction or guidance from their company. According to mobile application measurement company Flurry, each smartphone has, on average, 65 mobile apps installed. With 350 million employees expected to use smartphones at work by 2016, the number of mobile applications using the corporate network will continue to skyrocket.

The increase in mobile devices on the corporate network creates two distinct risks that lead to undesired or unpredictable network impacts. First, it contributes to the growing problem of unsanctioned applications on the corporate network.  Second, corporate policies cannot be consistently enforced across all devices.

Like web-based applications, the use of mobile applications is eroding IT control of the network. Unlike browser-based web applications, however, mobile applications and their corresponding operations are typically self-contained and fall outside the control of traditional web security solutions. Blue Coat says with these operational controls, businesses can set policies around specific functions within both web-based and mobile applications.

This enables businesses to enforce policy across all devices and allows them to regain control over the applications on their network. For example, a business that wants to mitigate unintentional leaks of confidential or secure location information could set a policy for Twitter that prevents employees from sending tweets but allows full reading or monitoring access.

“The growth of BYOD initiatives has created a situation where IT security managers are facing a deluge of untrusted, unmanaged devices and applications on the corporate network,” says Dave Ewart, director of product marketing, EMEA at Blue Coat Systems.

----Blue Coat

Read More

Power Firms Delving into Data

Details

Published on Sunday, 26 May 2013 05:02

A power cut in most places will bring down the street lights for a while. But in India last July blackouts nearly brought down the country. Around 700 million people were without power because of a grid collapse dubbed ‘Blackout Tuesday.' The power outage stretched around 2000 miles (3200 kilometers) across the north of the country, from Arunachal Pradesh in the east to Rajasthan in the west.

The disaster, like major outages previously in Paraguay, Java, and America, highlighted the frailty of power grids and put the spotlight on how utilities could avoid such events in future. The answer, many believe, is smart grids: power networks with embedded intelligence and a bigger capacity to deal with highs and lows in energy demand and supply.

Smart grids are deemed to be so critical to our energy future that in many countries there is a legal requirement for them to be introduced within a given timeframe. The data they supply, from advanced metering infrastructure (AMI) to distribution automation devices, can help utilities make more informed choices about how and where to generate and deliver power and how much to charge or pay for it. But this data poses a challenge for utilities, because there is so much of it. Traditionally, a power company might take a meter reading from a customer once a month to collect data. With smart grids, the amount of data has skyrocketed. Some utilities admit they have not yet really worked out what to do with all this information.

"It is not unusual for power companies to still be managing data at a level of granularity comparable to that of pre-smart grid days, despite now having technology that can deliver much more detailed information," says Smart Grid Update in its Data Management for Utilities briefing. However, it adds: "While the challenges associated with managing the masses of data associated with smart grids are certainly significant, they are not overwhelming. "Indeed, utility information management is simply one of several ‘big data' areas which a growing number of IT vendors are rushing to solve. The tools to deal with AMI data may yet need some refining, but they certainly exist."

So how can data analytics help power companies? They can help cut blackouts in two ways. First, having intelligence on the grid can help utilities catch failures as they are beginning to happen, and come up with workarounds that stop them from spreading. Second, a better understanding of grid behavior can help a utility balance supply and demand so the conditions for a blackout are less likely to happen in the first place. There are several technology companies that have developed advanced analytics systems for utilities, but for some utilities the problem is not what systems they need to buy to deal with smart grid data, but what changes they need to make to their own businesses.

"The utility is being inundated with phenomenal amounts of data," says Brian Rich, Vice President of Business Technology at Pacific Gas and Electric Company (PG&E), one of the largest combination natural gas and electric utilities in the United States. "Process analytics means that for the first time the company is breaking down traditional silos that haven't had the information required to integrate their processes." PG&E has so much smart grid information (70 terabytes of AMI interval data alone, increasing by 3 terabytes a month) that it has decided to replicate its data sets so it can perform deep analytics on the replicated set without interrupting the operational criticality of the data. This may sound like a lot of effort, but Rich says it is paying off for the utility.

"Before we were flying blind," he states. "Now we can have much more meaningful conversations with our customers about demand-side management programs." Plus the technology is helping PG&E to reduce grid failures. "We do a lot of over-the-air remediation," Rich explains. "If we have a meter not giving us readings we would previously have had to roll a truck. Now we can fix it by pushing a firmware upgrade over the network. "This, coupled with our significant investment in fault location, isolation, and service restoration, has enabled us to be more proactive about asset failures."

Ultimately, such measures may help make blackouts much more a thing of the past. Ben Kellison, a smart grid analyst at GTM Research, says: "Analytics and processing power on the grid is beginning to allow utilities to identify and isolate the causes of major blackouts before they cascade across the transmission grid. Meanwhile last-mile grid devices are providing the insight to reduce local outage duration, improve customer service, and increase a utility's operational efficiency."

By Jason Deign

The contents or opinions in this feature are independent and may not necessarily represent the views of Cisco. They are offered in an effort to encourage continuing conversations on a broad range of innovative technology subjects. We welcome your comments and engagement.

Used with the permission ofhttp://thenetwork.cisco.com/.

Read More

Proof of Concept

Details
    Published on Thursday, 05 January 2012 16:29
    Written by Craig Sutherland



At some point you will want to start the process of evaluating NMS suppliers and equipment. The proof of concept (PoC) is an important phase of the NMS evolution process. Sometimes it is difficult to be impartial during this process. Organisations often have a long standing relationship with one of the vendors involved in the PoC process, and they may have intimate knowledge of your network and its challenges.



To get the most from a PoC, you need to furnish each perspective vendor with as much information about your network and application as possible. In many cases, you may even find that a vendor withdraws from the evaluation process because its solution does not meet the PoC criteria. The following template can be used to send to perspective equipment vendors before the PoC starts. You will need to adjust the PoC criteria section to suit your own need. 



However, I have included some relevant sections which are generic and should help to get you started. Another suggestion with the PoC criteria is to invite each vendor to write the PoC criteria based on the information you provide in the template document. This way, you can compile the best questions from each vendor into a single criteria document, which is used as a common template document for all vendors.



Download pdf of the template here Proof of Concept template

Sample PoC Template Starts here.



    Background Information and Company Profile

XYZ corporation manufactures/provides [description of what the organisation does]. Each perspective vendor is invited by means of a PoC to demonstrate their proposed solution fills the business and technology requirements required to deliver a comprehensive network [performance] management solution.



The organisation operates in a [hub-and-spoke/distributed architecture] with a data distribution centre[s] located at [name of site]. Redundancy is achieved by [Hot Standby Core switching architecture/multiple redundant WAN links/Fully meshed MPLS topology]. The core switching infrastructure is running at version [enter version release] and the routing infrastructure is running at [enter version release].



CPU and memory utilisation on the network devices is generally[Low/High/Critical]. The organisation operates [Monday to Friday/24hrs] and network management operations are [continuously staffed by the operations centre/are required continuously during hours of operation and support staff are on standby outside of these hours].



    Duration of the PoC

The XYZ PoC evaluation process will last for [1 week/3 weeks]. Central management equipment will be hosted at [name of site] which has full management access to the network. To minimise disruption, vendors are requested to install software off site as much as possible. Vendors are asked to complete the following form:



 Duration

Please specify the amount of time required to set up the equipment and configure the system to achieve the requirements outlined in section 4 (Evaluation Criteria)? ___________



Space

Please specify the rack space required to house the equipment in RU units. In addition, please specify the depth of the equipment and power requirements. _______________



Network Map

The vendor is requested to submit a network map [included in this document] showing the placement of its equipment and the interaction if any with other components on the network.



Addressing

XYZ corporation operates a [IPV4/IPV6] addressing scheme and has allocated the following subnet for use during the PoC evaluation. [Subnet Range included here with network mask and gateway]



Connectivity

The equipment will connect to [Our live Network/Our test network] and as such, the vendor is required to submit the ports and port ranges and protocol types required for connecting to the network. Where not well known protocols are used, the vendors is required to provide an explanation of the ports' requirement and purpose it serves.

Ports and Protocols required for PoC evaluation: ________________________



Access

The vendor should use the following user name/password combination for accessing the system if multiple user name password combinations are used.

User name: [User name_Here]

Password: [Password_Here]



Configuration Change

If integrating the solution into the XYZ network requires modification to the existing network's routers or switches. The vendor should detail the changes needed to the current configuration to accommodate the monitoring solution. XYZ will then assess the changes requested and may recommend changes which are in line with the organisation's [security/IT] policy.



Anti-Virus

The vendor is required to specify the primary operating system proposed for this evaluation. Servers configured with Microsoft Windows [must have anti-virus software provided by XYZ installed and run the current recommended service pack. It should also/should have up to date virus software installed and the latest service pack installed] before any connection to the XYZ network is possible. Vendors should provide details of operating system hardening (if any) which is provided by default.



PoC Setup Form [Ends Here]



    Purpose of the PoC

    XYZ Corporation is in the process of selecting a solution and a partner to provide a [performance/management solution] for [which part of the network does this cover/the whole network] the purpose of the PoC is as follows:

    Demonstrate how the proposed solution integrates into the existing network without disruption to day to day activities.

    Meets the objectives outlined in the evaluation criteria.

    Copes with the volume of traffic currently present on the network and has the ability to scale to meet the future demand.

    Intuitive interface, should be easily accessible to the XYZ staff.

    Understand limitations of the proposed solution within the scope of the PoC.

    Evaluate all features to understand cost plus benefit.

    

    Evaluation Criteria

 The evaluation criteria will be a four-stage process and is outlined as follows:

    Ease of set up and connection to the network. [Score Maximum 25 pts]

    Ease of use, logical work flow, context sensitive help [Score Maximum 25 pts]

    Ability to meets or exceed technical requirements [Score Maximum 25 pts]

    Additional features, cost plus benefit [Score Maximum 25 pts]

    

The vendor shall generate or capture screen shots for functionality assessment for each of the questions outlined in the evaluation criteria. This will be included as [hard copy/soft copy] at the PoC wrap-up meeting.



 [Evaluation Criteria Starts Here]

    Access

    The system was accessible from all areas of the network within the scope of the PoC and functionality was not limited by firewalls. Changing the user name and password for the default access account and adding additional accounts with less access worked. Concurrent user access worked, how many concurrent logins were demonstrated.

    

    Compatibility

    The proposed system should be compatible with the existing routing and switching infrastructure in the network. Any issues such as software versions and memory should be highlighted.

    

    Performance and Accuracy

The accuracy proposed in the solution should be stated. Please state factors which may affect the accuracy of the proposed solution where applicable. Any shortcomings introduced by the scope of the PoC should also included.  If any performance related issues were discovered during the PoC phase such as, additional load placed on routing and switching components, delay and packet loss as a result of the solution integration, this shall also be stated.



    Reporting

    A list of interesting reports generated should be included in the PoC. This will include the report name and include a sample where applicable. The system shall also state the limitation in terms of custom report generation based on user defined fields and which custom reports worked well in this environment. Reporting generation- what was the most effective means to generate reports, scheduled or Ad-hoc, manually or via third party solution.

    

  [Evaluation Criteria Ends Here]



    Sample Reports

Sample reports should be taken during the peak hours of traffic activity. It should clearly indicate the name of the report and the link or network segment the report was taken from and the time interval for the report. Reports should be made in portable document format (PDF) as a soft copy where possible.



    Revised Bill Of Materials

 If for some reason during the PoC evaluation it was determined that:

    The evaluation equipment did not have enough capacity/storage/processing power to meet the objectives in the PoC criteria, the vendor is invited to submit a revised bill of materials which covers the unforeseen demands of the network or an alternative solution.

    

    To meet the objectives of the PoC criteria, if additional equipment is required either from the vendor or a third party, then the revised bill of materials should include this additional inventory.

    

    Additional features are not realised in the original PoC criteria, then the vendor should include this in the revised bill of materials.

Read More

Advanced Analytics a Priority

Details
    Published on Sunday, 26 October 2014 06:05

Advanced analytics is a top business priority, fuelled by the need to make advanced analysis accessible to more users and broaden the insight into the business, according to market analyst Gartner. Advanced analytics is the fastest-growing segment of the business intelligence (BI) and analytics software market and surpassed US $1 billion in 2013. "While advanced analytics have existed for over 20 years, big data has accelerated interest in the market and its position in the business," says Alexander Linden, research director at Gartner. "Rather than being the domain of a few select groups (for example, marketing, risk), many more business functions now have a legitimate interest in this capability to help foster better decision making and improved business outcomes." says Linden.



IT and business leaders must expand their efforts to move their organizations from using only traditional BI that addresses descriptive analysis (what happened) to advanced analytics, which complements by answering the "why," the "what will happen," and "how we can address it." While basic analytics provide a general summary of data, advanced analytics deliver deeper data knowledge and granular data analysis. The rewards of data-driven decision making can be a powerful boost to business outcomes. For example, clinical diagnosis applications apply analytics across a range of structured and unstructured sources — including electronic health records data, drug information, study data, and others — to recommend optimal treatment plans for individual patients.



However, creating value from data requires a range of talents, from data integration and preparation, to architecting specialized computing/database environments, to data mining and intelligent algorithms. "Extracting value out of data is not a trivial task," adds Linden. "One of the key elements of any such 'making sense out of data' program is the people, who must have the right skills and capabilities." Data scientists in the organization support big data initiatives for which the No. 1 use case is enhancing customer experience. According to Gartner's new global big data survey, 68 percent of respondents said that they use big data to enhance their customer experience. This is the third year customer experience has been a top business problem to address.



"Data scientists are not business analysts," says Linden. "They are professionals with the capability to derive mathematical models from data to reap clear and hard-hitting business benefits. They need to network well across different business units and work at the intersection of business goals, constraints, processes, available data and analytical possibilities."



Ultimately, data science is inevitable as it can help extract various kinds of knowledge from data, for example, how to acquire new customers (database marketing), how to do more cross-selling (via propensity-to-purchase modeling), information on route optimization, drug design and demand or failure prediction. "Whether an organization calls the role "data scientist" or something else, individuals (or teams of professionals) with these core skills and soft skills will prove essential in maximizing the realized value of your information assets, and discovering opportunities for enhanced business performance and competitive advantage," explains Linden.



--- Gartner

Read More