Thursday, November 30, 2006

More On How To Choose A Data Center

As a follow-up to an earlier article "How To Choose A Data Center" are just a few other considerations / concerns I would recommend. Combine the lists from both articles and you're more likely to make a smart decision choosing a data center.

Current Customers:

This is not a guaranteed seal of approval, but a Data Center's current customer base should give you an idea of the type of due diligence that others have performed on the Data Center you are evaluating. If numerous Fortune 50 companies are at a Data Center then there is a good chance it is a reliable place to keep your servers. NOTE: Some Data Centers will take on a "marquee" client at a loss in order to get the brand name. Make sure there are at least 2 high profile companies colocated there.

Employee Retention:

Ask the company how long the facility engineers have worked for that particular Data Center. High turn-over can be an indication of a number of things (including issues).

Power Outages:

Ask about the last power issue / outage they suffered - time-frame, length of the outage, reason(s) for the outage, mitigation steps. Most Data Centers will face some time of issue over the course of its use and you want to make sure they are honest with you about past issues, and have taken steps to correct any errors.

Bandwidth Connectivity:

How many "separate" bandwidth feeds are coming into the facility? A lot of companies say they have peering with "numerous" providers. BUT, all of that fiber may be coming into the building in single conduit. That means that if someone is digging a trench your Internet access may be cut off.

A top resource for determining your own business bandwidth requirements (T1, DS3, OC3, etc.) no cost to the the free services of

Data Centers requiring the most cost effective bandwidth solutions available at their location (OC3, OC12, OC48, OC192, etc.)....should take advantage of the free consultation and support from Data Center Bandwidth


If your company makes it big, will you have room to grow? Right now there are a LOT of Data Centers out of space and / or power. Make sure you aren't in a position where you have to choose another place to go down the road.

SLAs (Service Level Agreement):

Although there is no compensation for your e-commerce site taking a hit for an hour, make sure there are strict SLAs in place which will bring some financial relief in the case of an outage.

Managed Hosting:

There is a trade-off between owning the equipment (and keeping it at a Data Center), and utilizing a managed hosting provider. Make sure to analyze the pros and cons of both based on the current and future needs of your business. If you have older equipment, don't want to staff-up, I.T. isn't your core competence, etc, then consider going managed.

Data Recovery:

No matter where you go...make sure you have a DR plan in place. Keep backups of everything off-site and have a plan of action (hot-spares, DNS change-over, contact list, etc).

Fire suppression systems:

What provisions does the facility have for fire suppression? Is it a regular sprinkler system, Dry Pip Pre-action, gaseous CO2 or something else.

A Water based system is to protect the building, it will not protect the systems or the data because you're spraying water on computer systems in the event of a fire. A Pre-action system with dry pipes that only charge when there is a fire detected (smoke and heat detectors) is better because it prevents a construction accident where a sprinkler head is bumped or a pipe is broken from causing an incident.

Gaseous CO2, Energen, Halotron, Halon, etc are best, but more expensive. A combination of CO2 where power is run (under the floor perhaps?) and Pre-action sprinkler's above allows for a small fire to be contained in the underfloor space before water enters into the equation.

One thing to be very wary are what kind of hand held extinguishers are in the Data Center space? Are they Dry Chemical? If so, that's bad. Very bad. A fire (say in a garbage can) that would not jeopardize the data center or building can be made worse (in terms of the protection of the data center) because when someone goes to fight that fire, they're going to perhaps put the fire out (likely if small) but very likely spread a corrosive dust around the data center which will require either replacement of the systems OR professional cleaning of them. Look for halogenated or CO2 type extinguishers in the data-center for handheld use.

Physical Security:

Armed security guards are better than not armed in my opinion. Better training (ie more) and a more serious perspective on security of the building and the people. Guards more able to respond to serious problems are preferable to an old retiring that's more of a night watchman who calls 911 when there's an issue.

Physical Structure security is also important. What kind of walls does the building have? Are they several layers of concrete block or just one layer? Is it glass? Bullet Resistant glass (also good in storms) and is there an inner layer between the glass and the rest of the facility?

Do they have spill protection kits for dealing with water leaks? If there's a major rain storm and the roof is damaged, how does this impact the facility? Badly, slightly, total shutdown? Look for drip pans over critical infrastructure like PDUs and UPS systems.


Does the facility have a good loading dock that's near the data center or sufficiently proximate? If not, you're going to have a harder time getting your systems to the data center to install them. A good sized door large enough for the largest of racks and gear is important. Is there a lab where you can diagnose problems with your systems there? An area where you can stand up your systems if you need to? How are you getting final configuration work done at the site?

Generator Power:

Fuel Supplies cannot be underscored. A large on-site bunker is a good thing to see. A contract with a fuel supplier to be able to provide supplemental fuel on a daily basis is also important. I've seen an instance where a call center was left without power because the backup generator ran out of fuel after running for 24 hours. No-one thought o check the generator and get a fuel shipment to the call center.

Generator Maintenance can be important as well. A periodic maintenance contract with a major service company is good. But having someone that rides herd on the generator and power systems themselves is also important. Many small things that would not be noticed by a rotating batch of mechanics would be noticed by someone that can get a feel for all of the systems and perform periodic checks. It could be as simple as noticing that Generator No 2 sounds off and realizing there's an exhaust leak or as complex as a lubrication problem that is only found by periodic oil and coolant analysis by a lab that specializes in such tasks.

Periodic maintenance of the fuel is also important. Fuel that becomes contaminated with water can stop a generator system cold if bacteria begin growing in the fuel. The sludge that the bacteria produce will clog injectors and pumps as well as the fuel filters themselves. Periodic checks of the fuel as well as polishing and treatment is important.

There you go. Now you're fully armed with all the details you need to select just the right data center to meet your needs....and expectations.

Labels: , ,

Monday, November 27, 2006

The Basics Of How SIP (Session Initiation Protocol) Drives Your VoIP System Design And Function

Hold on to your seat.'s an explanation of SIP (Session Initiation Protocol) in laymans terms. It's pretty basic.... but should be just what you need to understand how this VoIP application fits into your VoIP system design plans and function. Afterall, not everyone in the business world can talk or understand "techie speak".

SIP - Session Initiation Protocol. It is exactly that - the main purpose of SIP is to set up and tear down media (audio/video/etc data) sessions, and also to manage endpoints and other things.

SIP devices communicate (usually) on UDP port 5060. When one device wants to start a call to another, it sends an INVITE message. Included in this is the SDP, session description protocol, which explains exactly what form the data will take (audio/video/etc, what codec, etc). When they agree and are ready to start exchanging media (data), RTP (realtime transport protocol) is used to actually exchange the data. RTP functions on any range of ports, which are assigned to each endpoint. The endpoints negotiate and choose acceptable ports on each side.

Sip also does a few other thing such as REGISTER. Register allows a SIP device with a dynamic IP to recieve calls. A common use is an ATA (vonage type box) - when you plug it in, it registers to its server, and renews the registration every XXX seconds to keep the server up to date (in case its IP changes).

SIP has a handful of other functions. For example NOTIFY can be used to pass assorted data to an endpoint (many IP phones will reboot when you NOTIFY them with the data 'check-sync'). NOTIFY is also used for MWI. There is also SUBSCRIBE, which allows an extension to subscribe to notifications of the status of a voice mail box (for MWI) or an extension/channel (for BLF....(busy lamp field, the thing which makes a guy's button light up when he is on the phone).

There are a handful of other SIP functions, for example REFER (transfer), BYE (hangup), etc.

SIP has three ways of dealing with DTMF signals sent while the call is in progress:

* Inband- Send the tones as audio in the media stream. Only works with G.711 ulaw/alaw codec, other codecs will distort the DTMF.

* RFC2833- Send the tones out of band but still attached to the audio stream via RTP.

* INFO- send the tones as SIP INFO packets along the control channel.

RFC2833 is probably the most common.

There is also a set of extensions called SIMPLE (Sip Instant Messaging, Presence, and Location Extensions). Put simply, this is a way to use SIP for instant messaging type uses.

SIP does not play nice with NAT routers, mostly because of RTP - the SDP includes the source and destination IP addresses where media should be sent to, which are not always correct.

For example - if you have an ATA behind NAT, it will use its own IP (192.168) when creating the SDP. NAT will correctly translate the header, so the packet is addressed from the network's external IP. However the contents of the packet still have a 192.168 IP as the destination, which the server cannot send media to. This commonly results in calls which work except one or both parties cannot hear each other.

There are two ways to solve this - media gateways (sip-aware router that rewrites the SDP) or more commonly, STUN (Sip Traversal Under NAT). STUN is a protocol which allows a SIP device to, with the help of a STUN server, discover its own external IP and what kind of NAT it is behind. It can then write the SDP correctly and negotiate the RTP session so the NAT will not bother it.

SIP shares many response codes with HTTP. IE- 404=extension not found, 401=unauthorized, etc. case you are ever looking at a SIP transcript - SIP performs authentication (where passwords are used) using digests. Thus, a typical authenticated session looks like this:

device tries to connect (INVITE)....
server responds trying....
server responds 401 unauthorized with some auth info....
device responds OK....
device tries to connect (INVITE) this time with hashed auth data....
server responds trying....
server responds ok (and other phone starts to ring)....

Hopefully the above gives you a basic understanding and arms you with enough to be able to ask the right questions....for the right the right time.

Now that you have an understanding of SIP....when you're ready to implement a business VoIP system we recommend using this free consulting service to find the most cost effective solution: Business VoIP

Thursday, November 23, 2006

Gigabit Ethernet ....Business Solution For Increased Network Performance

In a scenario all too familiar to network managers, users begin complaining of slower response times. Network management tools indicate that a high-transaction server is hogging bandwidth in a shared domain or that server CPU utilization is not optimal – classic bottleneck symptoms. What’s the best solution? More and more often, IT managers are turning to Gigabit Ethernet technology.

When Gigabit Ethernet was standardized for fiber optic cabling in the late 1990s, IT managers began to see the benefits of Gigabit speeds applied to the network backbone and in the data center. Today, with Gigabit over copper, 1000Mbps connectivity is being deployed even more widely – including switch stacks, servers and desktops – using the cabling that is already in place in most buildings.

IT managers have found that Ethernet is simple, easy to use and readily upgradeable. An organization can scale from 10 to 100 or 1000Mbps Ethernet, either network-wide or a segment at a time, knowing that the new equipment will be backwards compatible with legacy equipment. This reduces the infrastructure investment that an organization must make. Ethernet is also a reliable technology. Experience shows that it can be deployed with confidence for mission-critical applications.

A prime reason for the success of Ethernet is the fact that Ethernet standards, over 25 years old, have progressed along with networking requirements. This progression of standards provides a clear and straightforward migration path for companies as their bandwidth requirements increase.

In general, upgrading to Gigabit Ethernet over copper is ideal even when budgets are tight, because it increases performance by building on a company’s current investment in cabling. Typically, the network infrastructure simply adjusts to this higher speed. No technician re-training is needed, and any disruption to the network is usually minimal. What’s more, Gigabit over copper is the most economical cabling choice in terms of cost per Mbps.

Gigabit Ethernet is also a good choice because it supports Quality of Service (QoS) methods that are increasingly important for avoiding latency problems as voice, video and data share the cable for Next-Generation Networking (NGN) applications. Like Fast Ethernet, Gigabit Ethernet supports existing traffic management techniques that deliver Quality of Service over Ethernet, such as IEEE 802.1p traffic prioritization and Multi Protocol Label Switching (MPLS).

Benefits of Gigabit

Gigabit Ethernet is 100 times faster than regular 10Mbps Ethernet and 10 times faster than 100Mbps Fast Ethernet. The principal benefits of Gigabit Ethernet include:

* Increased bandwidth for higher performance and elimination of bottlenecks

* Full-duplex capacity, allowing the effective bandwidth to be virtually doubled

* Aggregating bandwidth to multi-Gigabit speeds using Gigabit server adapters and switches

* Quality of Service (QoS) features to help eliminate jittery video or distorted audio

* Low cost of acquisition and ownership

* Full compatibility with the large installed base of Ethernet and Fast Ethernet nodes

* Transferring large amounts of data across a network quickly

Deployment of bandwidth-hungry applications, more powerful processors and increasingly powerful operating systems are likely to continue for the foreseeable future, requiring faster network connections. Just as networking has historically moved to higher-speed connections in stages, individual networks also proceed from early to mainstream adoption.

First, network managers must think through the network environment to identify the power users/segments of their network that would need highperformance connections first. These users or segments are often retrofitted with Gigabit connections without waiting for the normal cycle of PC replacement.

Network managers then begin purchasing Gigabit-enabled systems for other segments in the organization. The most cost-effective method is to order new PCs equipped with triple-speed 10/100/1000Mbps connections to prepare these desktops for the coming increases in bandwidth requirements. Organizations want to avoid being “blind-sided” by the new wave of next generation applications.

To help you navigate the network design issues....and obtain bandwidth rate quotes from multiple providers available in/at your install locations...I strongly suggest using this free service: Gigabit Ethernet

Monday, November 20, 2006

Burstable DS3 Bandwidth....Smart Network Solution In The Right Situation

In today’s business environment, Internet bandwidth requirements are constantly increasing to keep up with demand — everything from end-user Internet accessibility to the sharing of enterprise applications. In addition, there is access to corporate intranet or extranet resources, Internet-based e-mail services, large data and file transfers between networks, and advanced broadband applications such as streaming media, multi-media, video conferencing, and more.

If you require full-time, super-high-capacity access to the Internet, with moderate bandwidth usage most of the time, but very high bandwidth usage in short bursts, your business WAN or LAN solution may benefit best from a burstable DS3 Internet connection.

This connection provides your business or enterprise with a dedicated, flexible, 24-hour-a-day high capacity, low latancy circuit to the Internet via multiple Tier I peering arrangements. You normally receive 4 Mbps minimum Internet access with a maximum of 45 Mbps...often billed in 1 Mbp increments after the first 4 Mbps. Therefore fixed customer fees are based on the monthly commitment, leaving open the option for bursting whenever your traffic load requires. You use what you need....when you need it....and only pay for what you use.

Now that makes business sense.

To request free rate quotes for burstable or full DS3 bandwidth service.... comparing multiple providers available in your area....use this free service:

DS3 Bandwidth

Thursday, November 16, 2006

How Does A T1 PRI Work?

Here's a brief description of the difference between T1 voice vs PRI.....

"PRI or Primary Rate Interface is a switched service deliver OVER a T1 connection . If someone were to say “I want to order a PRI to location X.” What is being requested is a connection to a trunk side module capable of ISDN protocol to a Telco switch delivered to location X via a DS1 rate (T1) circuit.

Now if someone were to say “I want to order a T1 to that SAME location X.” … well… That wouldn’t be enough info… A T1 from where to where? A T1 for what? By just asking for a T1, nothing is understood or implied about where or what the circuit would be used for. A T1 can be used to truck data at the rate of 1.544 Mbps from one location to another… With channel banks and appropriate DS0 level cards, a T1 can be used to truck up to 24 separate and distinct DS0 signals (analog data, analog voice, or Digital Data Service) from one location to another… A T1 can be used to connect one location to an ISP Internet Edge device to connect a customer to the World Wide Web… In other words, a T1 is a multi use pipe…"

With a point to point T1 you can use it for voice and/or data depending on the equipment you have at each end. They are basically just providing the "Pipe" and it is up to you what it is used for.

The thing with T1's and PRI's is it all depends on what equipment is connected to it on each end.

As quoted above, a T1 is just a 24 channel circuit that can be used for multiple things. A PRI is a protocol that uses the 24th channel to control what the other 23 channels are used for.

In a nutshell - a T1 has 24 channels using 56K for data and 8K for signalling. A PRI has 24 channels and uses 23 for data and the 24th for signalling. Also you can break down the channels on a T1 for different type of service. On the PRI all 23 channels have to be the same.

The features... or lack of... totally depend on where the circuit is going and what it is connected to on the other end.

Keeping it on simple terms...

PRI is a diffrent protocol to be used with voice services over the T1 line. Where as a t1 line can also be used for voice but without using the PRI protocol and probably using diffrent equipment at the other end to get the dial tone....

Of course....there are a lot more technical details dealing with ISDN lines (PRI's/BRI's) but you've got the basic idea.

To request quotes for PRI and/or T1 voice service.... comparing providers available in your area....use this free service: T1 Voice and PRI Rate Quotes

Monday, November 13, 2006

DSL Installation Tutorial .... Everything You Need To Know To Install DSL Internet Service

If you're having a hard time getting quality DSL speed because of poor line quality, take a second to read the below article. It proposes a couple of solutions that are helpful for resolving most problems you may encounter.

Learn To Install DSL

Per the above article.....if line quality is an issue I'd suggest getting the SPSDL0SR1 DSL Splitter Module described at the link below. If you end up going that route:

DSL Splitter Module

Where you may be searching for another DSL provider this simple tool will compare rates from multiple providers for you (US only....and will also cover cable and satellite providers also): Compare DSL Rates

Thursday, November 09, 2006

Understanding MPLS (Multi-Protocol Label Switching)

In laymen’s terms..... what is MPLS? How does it work? What are the advantages?

MPLS provides a fully meshed private IP network that can replace legacy networks, such as Frame Relay / ATM, and are specifically engineered for next generation IP applications including VoIP and video. MPLS is native IP, it might go over a frame/ATM local loop for last mile connectivity, but the core network will ride over POS links or better.

MPLS can reduce total telecom costs by as much as 70% by combining voice-data-video on one network, elminate ALL intra office long distance costs, secure your data using VPN, has built in disaster recovery with it's fully meshed network.

Basically it's the future of WANs as everything migrates to a more robust, scalable, less costly IP based solution.

If you have any specific questions....and need rate quotes comparing available MPLS providers in your should use this free service: MPLS Network

Monday, November 06, 2006

Just What Is A Bonded T1 And What Does It Give You?

A bonded T1 just means that the you have two T1 lines (from the same provider) which have been joined together in a special way so that you can use the combined total of 3.0 Mbps of Internet data or the 46 channels for voice/phone use. You cannot bond T1's together if they are not through the same provider.

If you were to have two T1's, unbonded, then you could not use the full 3.0 Mbps of bandwidth all at once. For example, if you did a speed test, you wouldn't be able to get a result of "3.0 Mbps", the max you could download/upload at any one time would be 1.5 Mbps, even though you had two T1's. I know it doesn't make sense because you think that 1.5+1.5=3.0. It doesn't work like that. Unless your T's are bonded, you can only use the 1.5 Mbps of bandwidth that each line provides....seperately. You can do two download tests, at separate download test sites, and get a full 1.5 Mbps reading at each download site. But with out bonding, there isn't a way of getting a test result of 3.0 Mbps. Think of it in terms of cars: You can either have two slow Datsun pickups or one Ram 4x4 with a 5.7L Hemi. Sometimes, two slow pickups are better for reliability than one big Ram 4x4, but not as big and fast.

There are exceptions to this of course, and there's probally better ways of explaining it, but this will give you a general idea.

You can bond quite a few T1 together by the way, you're not limited to just two. The maximum is usually around 4 T1's although some have gone as far as 8. However, if you go beyond 4 bonded T1's it makes much more sense to start looking at a fractional/burstable DS3.

A simple tool to request quotes for bonded T1's from multiple providers available in your location can be found here: Bonded T1

If you need more than what a bonded T1 can provide...such as a fractional or full can request quotes for DS3 Bandwidth from multiple providers here: DS3 Bandwidth

Thursday, November 02, 2006

Definitions And Background Of Telecom And Internet Service Terms

Below is a fairly comprehensive a list of Telecom and Internet Service Terms with definitions and/or background info for each. Hopefully you'll gain the necessary understanding you need for whatever your application.


10 Mbps baseband Ethernet specification using 50 ohm thin coaxial cable. Also known as Thin Ethernet.


100 Mbps baseband Fast Ethernet specification using two strands of multimode fibre-optic cable per link. To guarantee proper signal timing, a 100BaseFX link cannot exceed 1312 feet (400 meters) in length.


10 Mbps baseband Ethernet specification using two pairs of twisted-pair cabling.


100 Mbps baseband Fast Ethernet specification using UTP wiring. Like the 10BaseT technology on which it is based, 100BaseT sends link pulses over the network segment when no traffic is present. However, these link pulses contain more information than those used in 10BaseT.



Add/Drop Multiplexer. A device which inserts lower rate channel traffic [add] or removes the traffic [drop] from a higher rate aggregate channel in a synchronous transmission network (SDH or SONET). Typically adds or drops traffic at STM-1 or higher from a STM-64 aggregate signal.


Asymmetric Digital Subscriber Loop. A family of technologies used to deliver high-rate digital data over the existing copper local loop. One-way rates of up to 6 Mbps downstream (from the central office to the subscriber) and up to 640 kbps upstream have been achieved, although the technology is still evolving and higher data rates are expected in the future.

ADSL is expected to grow to become the main connection to the Net for the home and small business user due to the relatively low cost of the technology.


American National Standards Institute.

Application Layer

Seventh (top) layer in the OSI seven layer model. Layer Contains functions for particular applications services such as file transfer.


American Standard Code for Information Interchange. Method of encoding characters digitally.


Applications Service Provider. An organization which provides applications (word processing, accounting packages, e-business, etc) remotely over a network.


Asynchronous Transfer Mode. ATM is a connection-oriented packet switching technique where all cells follow the same path through the network. Cells are made up of 48 bytes of data plus a 5 byte header (53 bytes total). See for more details.


The percentage of time for which a circuit or system is
available for use. Carrier class is typically 99.999% - about five minutes downtime per year.



The high-speed - usually long-haul - central portions of a telecommunications network.


Describes the width of the pipe into the customer. ’Bandwidth on demand’ refers to the ability to increase the bandwidth as needed.


Baud - Unit of signaling speed in a communications system. Not to be confused with bps. One baud means one signal element transmitted per second.

A signal element may contain zero, one or more than one bit of information,depending on the method of modulation.


Border Gateway Protocol version 4. The major exterior gateway protocol used in the internet. Specified in RFC 1772.


BInary digiT. A 1 or a 0. The basic element of digital communication.


A specification for local area wireless communication. Bluetooth allows devices to discover each other and communicate without human intervention. See for more details.


Bit(s) per second. Also b/s. Transmission capacity of a channel or system.


Basic Rate Interface ISDN connection providing 2B+D, where B is a 64 kbps Bearer and D is a 16 kbps data channel.


Device that connects two or more networks together and forwards packets between them. The networks may use dissimilar protocols. A bridge operates at the data link layer (layer two) as opposed to a router which operates at layer three.


High-speed or high-capacity channel or system


8 bits of information.


Carrier Class

A statement of system or network availability. Carrier class is typically 99.999% - about five minutes of downtime per year.


Committed Information Rate. A specified amount of guaranteed bandwidth (measured in bits per second), usually on a Frame Relay service. Typically, when purchasing a Frame Relay service, a company can specify the CIR level they wish. The Frame Relay network vendor guarantees that frames not exceeding this level will be delivered. It's possible that additional traffic may also be delivered, but it's not guaranteed. Some Frame Relay vendors offer inexpensive services with a CIR equal to zero. This essentially means that the network will deliver as many frames as it can, but it doesn't guarantee any bandwidth level.


Customer Premises Equipment. This is the equipment at the customer site which terminates the connection into the network and connects to the LAN for private networks, Internet connectivity and voice services.


Dark Fiber

A fiber-optic strand with no optical transmission equipment. Customers add their own equipment and build their own network, retaining complete control over all aspects of it.


The basic unit of information passed across the Internet. It is a self-contained packet containing source and destination addresses as well as data. See also frame and packet.

Data Link Layer

Second layer in the OSI seven layer model. Layer Transmits packets from node to node.


deciBel. A logarithmic unit which defines the ratio between two powers P1 and P2. Ratio in dB = 10 log10 (P1/P2). The original unit was the Bel, named after Alexander Graham Bell, but is inconveniently large.

Used in power budgets to describe the optical loss over a particular link and therefore the optical power needed over the link in order to get a signal to the other end.

Used in fiber specifications to describe the loss per meter in the fibre, where the lower the loss the better.


Direct Broadcast Satellite - where the end-user receives the broadcast signal direct from the satellite. As opposed to a satellite feed to a cable TV head end, for subsequent distribution to end-users via the cable network.


Data Communication Network. Used to convey Network Management commands and reports around a communications network infrastructure.


Dynamic Host Configuration Protocol. A method of assigning an IP address dynamically to a device each time it connects to a network. DHCP simplifies network administration and means that normal PC's on a network do not need a fixed IP address.


Normally refers to chromatic Dispersion. Chromatic Dispersion is the variation of propagation velocity - speed of travel of the light - in an optical fiber with frequency - and with wavelength - causing the pulses of light to be degraded and merge into each other as the light travels down the fibre. See also PMD. Different types of fiber have different Dispersion characteristics. Management of Dispersion is key to network design at the fiber level.


Demilitarized Zone. Part of the network to which access is controlled by a Firewall.


Domain Name System. A distributed database which provides the mapping/translation between the domain name and the individual IP address allocated to that host.

Domain Name

The domain name is the unique name that identifies an organization on the Internet. The Fully Qualified Domain Name (FQDN) includes the hostname and is easier to use than the numerical IP addresses which are used to route traffic to the correct destination. This means that you only need to remember the name of the web site rather than its IP address.

A given server may have or support more than one domain name, but a given FQDN points to only one host.

com Commercial
edu Educational
gov Government (US unless specified)
int International
mil Military (US unless specified)
net Network
org Non-profit organization

Subsequently the system was extended to allow an optional final two-letter country group. Examples are given below.

au Australia it Italy
br Brazil jp Japan
ca Canada mx Mexico
de Germany nl Netherlands
es Spain no Norway
fi Finland se Sweden
gr Greece tr Turkey
hu Hungary uk United Kingdom
in India us United States

DS-0, DS-1

Digital Service.


Dispersion Shifted fiber (G.653). fiber manufactured to achieve zero chromatic dispersion at 1550 nm - the wavelength of lowest loss.


Dense Wavelength Division Multiplexing. Multiplexing of signals by transmitting them on typically eight or more different wavelengths on the same fibre. The ’dense’ prefix usually signifies systems where the spacing between wavelengths is smaller than the first generation WDM systems.



Describes the first level in the European digital hierarchy and is 30 channels of 64 kbps, plus two channels signaling and framing, to give an overall rate of 2.048 Mbps.


Erbium Doped fiber Amplifier. Optical fibres doped with the rare earth element erbium, which can amplify light in the wavelength region from 1530nm to 1625nm when pumped by an external light source. EDFAs allow an optical pulse to travel long distances before electrical regeneration is required.


A LAN access method defined in IEEE 802.3. It is a shared medium approach originally developed at Xerox Palo Alto Research Center using ideas from the University of Hawaii Aloha packet radio network. Available at 10 Mbps (10BaseT), 100 Mbps (100BaseT) and now 1 Gbps (Gigabit Ethernet).

Ethernet functions at layers one and two of the OSI model.

Also known as Carrier Sense Multiple Access with Collision Detection (CSMA/CD). Ethernet is the most common LAN mechanism within an organization See also Fast Ethernet.


European Telecommunications Standards Institute. Located at Sophia-Antipolis near Nice, France. Defines technical standards and interfaces in Europe. See also ANSI and ITU.


Fast Ethernet

A 100 Mbps version of 10 Mbps Ethernet.


Forward Error Correction. Technique for detecting and correcting errors (from imperfect transmission) by adding a small number of extra bits. FEC allows optical transmission over longer distances by correcting errors that can happen as the signal-to-noise ratio decreases with distance. See also Raman Amplification.

Fibre-optic Cable

Network cabling that employs one or more Optical Fibres.


Device placed between an organization’s private network and the Internet to authenticate incoming users. A Firewall is normally a specially configured computer which is set up to only allow specific incoming traffic and users onto the network.


1. A block of data in a specified format.
2. A rack housing telecommunications equipment.

Frame Relay

High-speed packet switching technique used to interconnect LANs. Capable of any payload up to 4096 bytes per packet. Defined in ITU-T I.122. Typically used to build VPN's, particularly where guarantees of bandwidth are required.


Four Wave Mixing. Describes the generation of unwanted sidebands when two or more high-power optical signals exist in the same non-ideal medium.

FWM is one of the elements that need to be taken into account when designing optical networks, and must be minimized if interference is to be avoided.


File Transfer Protocol. Operates at layers five, six and seven of the OSI model and allows log-on to a remote host, directory listing and file transfer.



109 bits.


The router which provides the connection between the LAN and the WAN.


Global Positioning System. Allows a GPS device to determine its location to better than 20m anywhere on, or above, the Earth by interpreting the signals from up to five orbiting satellites.


The process of combining partially filled trunks into a smaller number of fully filled trunks.


Global System for Mobile communications. Used by second generation mobile phones to connect to the mobile networks.

ITU-T recommendation



HyperText Markup Language. Simple hypertext document formatting language that uses tags to indicate how a given part of a document should be interpreted by a viewing application, such as a Web browser.


HyperText Transfer Protocol. Part of the TCP/IP suite. The underlying protocol of WORLd Wide Web pages, used to communicate between the user’s browser and the Web server.


Central connecting point for star-connected circuits. Active hubs contain regeneration facilities.



Internet Architecture Board. Board of internet work researchers who discuss issues pertinent to Internet architecture. Responsible for appointing a variety of Internet-related groups.


Internet Engineering Task Force. Task force consisting of over 80 working groups responsible for developing Internet standards (produced as RFC's). The IETF operates under the auspices of ISOC.


In-Line Amplifier.


The interconnection of computers across the world that evolved from the ARPAnet (Advanced Research Projects Agency).

The Internet is made up of many networks each run by a different company and interconnected at peering points. The common use of IP and Internet standards allows users connected to one network to communicate with users on another network.


Internet Protocol. Defines the unit of information passed between systems providing a basic packet delivery service within the TCP/IP. IP is a standard that describes how packets of data are transported across the Internet and recognized as incoming messages.

IP addresses/ addressing (IPv4)

The unique 32 bit address for a specific TCP/IP host in the Internet.


IPSec provides security for transmission of sensitive information over unprotected networks such as the Internet. IPSec acts at the network layer, protecting and authenticating IP packets between participating IPSec devices.


Latest version of IP ( IPv6 uses 128 bit address space compared with the 32 bit IPv4 address.


Indefeasible Right of Use. The provision of the Right of Use for a long period of time, usually 15+ years, of a bandwidth service - dark fibre, wavelengths or SDH - for an up front fee.


Integrated Services Digital Network. See also PRI.


International Standards Organization


Internet Society. International non-profit organization founded in 1992 to coordinate the evolution and use of the Internet. In addition, ISOC delegates authority to other groups related to the Internet, such as the IAB.


Internet Service Provider. Connects the end-user to the Internet.


International Telecommunications Union



Variation in timing, or time of arrival, of received signals; an unwanted lack of perfection which can lead to bit areas.



One thousand bits (103).



Local Area Network. Used within a building to link computers and other devices, such as printers. Typically uses Ethernet.


Time taken to deliver a packet from the source to the receiver. Includes propagation delay (the time taken for the electrical or optical signals to travel the distance between the two points) and processing delay. Due to the distance to a satellite and back (over 34,000km each way), the latency when communicating via a satellite connection is at least 270 milliseconds, making interactive services difficult, compared to a delay of about 10 milliseconds across Europe via fibre.

Local Loop

Originally, the pair of wires (loop) between the subscriber (to a telephone system) and the local telephone exchange (switch or office). Now used as a generic term to describe the connection between the last switch/routing point and the subscriber, no matter what technology is used to deliver the service, nor what service (voice, data etc.) is delivered.



Metropolitan Area Network.


One million bits (106 bits).


Mode Field Diameter of an optical fibre. The diameter at which the electric and magnetic field strengths are reduced to 1/e of their maximum values (for a Gaussian distribution in a single mode fibre, and where e is the base of natural logarithms, 2.71828...). This is the practical replacement for core diameter in single-mode fibre.


Management Information Base. Database of Network Management information that is used and maintained by a Network Management protocol such as SNMP or CMIP. The value of a MIB object can be changed or retrieved using SNMP or CMIP commands, usually through a GUI Network Management system. MIB objects are organized in a tree structure that includes public (standard) and private (proprietary) branches.


MOdulator/DEModulator. Device for converting data signals to/from forms suitable for transmission over an analogue voice channel.


MultiProtocol Label Switching. A short fixed-length label is generated that acts as a shorthand representation of an IP packet's header. Subsequent routing decisions (made by Label Switched routers) are made based on the MPLS label and not the original IP address. This new technology allows core network routers to operate at higher speeds without needing to examine each packet in detail, and allows more complex services to be developed, allowing discrimination on a QoS basis.


Multiplex Section


Multiplex Section Shared Protection Ring


Mean Time Between Failures. A key metric for the quality of equipment and a determining factor in the overall SLA that can be achieved.


Mean Time To Repair. The usual time taken to fix a problem that occurs on the network. Targets are normally set within an SLA and depend on the priority of the fault.



Network Address Translation. Used by a Firewall or Gateway to hide LAN IP addressing from devices on the WAN.


Non Dispersion Shifted fiber (G.652)

Network Layer

Third layer in the OSI seven layer model. Determines routes based on network address.

Network Management

The processes of managing, monitoring, and controlling a communications network. Modern Network Management systems also include the ability to re-configure network elements remotely.


Network Operation Center


nanometer. 10-9 meter


Non-Zero Dispersion Shifted fiber (G.655). Optical fiber which has been optimized for DWDM operation by having a small but finite amount of dispersion present at 1550nm.



Group of 8 data bits.


Optical Carrier - x.


Optical Distribution Frame.

This is the point where the optical fiber within the backbone network terminates and the customer's equipment of fiber connects, providing the demarcation point between i-21 and the customer.

Optical Fiber

A method of guiding light over long distances with very little reduction on strength (attenuation or loss). A central core of high-refractive index material - usually very pure glass - is covered with a cladding of lower refractive index material. Modern fibres have loss in the order of 0.25 dB/km, so 1 km of fiber has less loss than a pair of ordinary spectacles or reading glasses.


From Optoelectronics. The combination of optics and electronics.


Optical Return Loss. Ratio of power reflected (from a connector or other discontinuity) to incident power. Usually expressed in dB.


Open Systems Interconnection. An ISO standard defining a communications model with seven layers. The layers are: Layer 1 Physical layer - electrical or optical signals Layer 2 Data Link layer - transmits packets from node to node Layer 3 Network layer - determines routes based on network address Layer 4 Transport layer - manages end-to-end delivery, including flow control and error recovery Layer 5 Session layer - initiates and manages communications session Layer 6 Presentation layer - performs any necessary character code conversion to provide transparent communications Layer 7 Application layer - contains functions for particular applications services such as file transfer and file access.

More than one layer may be combined into one module or process.



A block of data. The terms packet, frame, and datagram are often used interchangeably.


Internet interconnection as equals (peers) and thus no billing between the parties. This is the normal method of interconnection between the sub-networks which make up the Internet. Contrasts with supplier/ customer interconnection, e.g. ISP and end customer.


1015 bits.

Physical Layer

First layer in the OSI seven layer model. The electrical or optical signals physically transported across the network.


Almost synchronous because bits are stuffed into the frames as padding and the call’s location varies slightly - jitters - from frame to frame.


Polarisation Mode Dispersion. Dispersion caused by different material properties for different planes of polarisation (direction of the electric field) in an optical fibre.


Point of Presence (or Access Node). A site where customers can connect into the backbone network.


Plain Old Telephone Service

Power Budget

The optical budget in dB over a specific network link. Usually quoted as an ’end-of-life’ figure which takes into account some margin for repairs and expected very small deterioration in the quality of the fiber over time.

Presentation Layer

Sixth layer in the OSI seven layer model. Performs any necessary character code conversion to provide transparent communications.


Primary Rate Interface. ISDN connection at E1 (30B + D at 2.048 Mbps) or T1 (23B + D at 1.544 Mbps) speed, where B is a 64 kbps Bearer and D is a 64 kbps data channel.



Quality of Service. A definition of the service provided to a customer.


Raman Amplification

A technique for amplifying optical signals in which high-power laser light is sent in the direction opposite that traveled by the data signals, transforming part of the transmission fiber into an amplifier of the signals passing through it. Raman Amplification is named after the scientist who discovered the phenomenon in the scattering of light, called the Raman Effect, in 1928. Typically used to extend the distance that optical signals can be transmitted. See also FEC.


Remote Dial-In User Service. Database for authenticating modem and ISDN connections and for tracking connection time.


Request For Comments. Document series used as the primary means for communicating information about the Internet. Some RFC's are designated by the IAB as Internet standards. Most RFC's document protocol specifications such as Telnet and FTP but some are humorous or historical. RFC's are available online from


Réseaux IP Européens. Group formed to coordinate IP based networks in Europe. This is the organization within Europe that allocates IP addresses to ISP's who in turn allocate them to customers.


Device connecting two or more networks together which forwards packets between them. Routers read the network address and use routing tables to find the best route between the networks. The routing tables can be created automatically by the system. Routers can also implement load balancing and generate statistics. A router operates at the network layer (layer three) as opposed to a bridge which operates at layer two.



Type of optical connector. Type SC, Super Physical Contact. The type SC connector was originally developed by NTT of Japan. The suffixes /PC, /SPC, /APC are terms which describe connector end-faces and also relate to the ORL designation.

PC means Physical Contact, a description of the contacting spherical end-face. PC has come to mean an ORL greater than 35db. SPC means Super PC, which means a PC connector with ORL >45db.

APC means Angled PC (the end face is polished at an angle, usually 8°) which improves ORL to >65db.


Synchronous Digital Hierarchy.

Session Layer

Fifth layer in the OSI seven layer model. Initiates and manages communications session.


Standard Generalized Markup Language. Describes the relationship between a document's contents and its structure in an open (not vendor-specific) format. SGML is defined in "ISO 8879:1986 information processing - text and office systems - Standard Generalized Markup Language (SGML)."


Le Système International (d’Unités). The International system (metric) of units of measure.


Switched Multimegabit Data Service


Simple Mail Transfer Protocol. Part of the TCP/IP suite. This is the protocol used for transporting email over the Internet, between email servers (which hold and store email) and clients (which allow users to read the email).


Simple Network Management Protocol. An application layer protocol that facilitates the exchange of management information between network devices. Part of the TCP/IP suite.


Synchronous Optical Network.


Term used to describe unsolicited email or newsgroup posts, often in the form of commercial announcements. The act of sending a Spam is known as Spamming.


1. A method of reducing network traffic by simulating local responses to 'keep alive' routine queries to distant devices. Typically used to conserve WAN bandwidth or to avoid unnecessary call set-up and resulting phone call charges.

2. A method of gaining unauthorized access to a system (hacking) by simulating the identity of a genuine user or of a trusted entity.


Secure SHell. Protocol or program for secure logon to a remote host over an insecure network.


Secure Socket Layer. Encryption technology for the Web used to provide secure transactions such as the transmission of credit card numbers for e-commerce.


Type of optical connector. The ’straight tip’ connector developed by AT&T, it features a physically contacting non-rotating 2.5mm ferrule design and bayonet connector-to-adapter mating.

STM-1, STM-x

Synchronous Transport Module - 1, x.


In a telecom's system synchronous means ’bits from one telephone call are always in the same location inside a digital transmission frame’.



Transmission Control Protocol/Internet Protocol


1012 bits.

Traffic Engineering

The process of determining the amount of network capacity (circuits, bandwidth) required to handle a specified amount of traffic without exceeding a specified number of lost calls or dropped packets.

Transport Layer

Fourth layer in the OSI seven layer model. Manages end-to-end delivery, including flow control and error recovery.

T-1, T-x

A T-1 system transports a DS-1 signal at 1.544 Mbps which comprises 24 DS-0 channels multiplexed together.


A multi-channel communication link such as an E-1 or T-1.



Universal Mobile Telephone Service. Third generation cellular phone technology.


Uniform Resource Locator. Formerly Universal Resource Locator. A form of host address used on the Internet - e.g. is the URL for the T1 Shopper web site


Unshielded Twisted Pair. Cabling with typically four twisted pairs (eight wires), used for cabling within a building. Typically used for 10BaseT cabling.


Virtual Private Network

Virtual Private Network. A network provided to the customer which is invisible to any other users of the backbone network.

This provides security to the customer (if the network cannot be seen it cannot be interfered with) and allows Quality of Service specific to the customer to be provided. VPN's allow the customer’s view of the network to be greatly simplified and tailored to specific requirements.


Voice over IP. The ability to carry Packetised voice over an IP-based Internet with POTS-like functionality, reliability and voice quality.



Wide Area Network. As distinct from Local Area Network. A WAN connects multiple LANs together. Typically an ISP or service provider provides the WAN into which a company will connect their LANs from each site.


Wireless Application Protocol. Delivers information and services to This is essentially a minimal form of the Web for mobile phones.


Wavelength Division Multiplexing. Generic term for the technique of simultaneously transmitting more than one wavelength of light down an optical fibre. Also see DWDM



eXtensible Markup Language. Text markup language designed to enable the use of SGML on the World-Wide Web. XML allows customized markup languages to be defined.

ITU-T recommendations s for data communications.



Cable Design

Fibre types SM - fiber in accordance with ITU-T G.652, G.653, G.654 or G.655

Number of fibres 96

Strength member ... 3.0 mm fiber reinforced polymer rod

Stranding and coding ... 6 loose tubes are stranded around the strength member. Each of the tubes contain up to 16 fibres

Filling ... Thixotropic jelly

Wrapping ... Double layer of polyester tape

Sheath Inner ... 1.0 mm polyethylene, 2.5 ± 0.5% carbon black

Sheath Middle ... 1.5 mm lead

Sheath Outer ... 1.8 mm polyethylene, 2.5 ± 0.5% carbon black

Armouring ... Bedding, polypropylene yarn, all bedded in bitumen. Steel wire, bedded and floated in bitumen. Double layer of polypropylene yarn, bedded in bitumen

ITU-T Recommendations

G series recommendation - Transmission systems and media, digital systems and networks

G.652 ... Characteristics of a single-mode optical fiber cable

G.653 ... Characteristics of a dispersion-shifted single-mode optical fiber cable

G.654 ... Characteristics of a cut-off shifted single- mode optical fiber cable

G.655 ... Characteristics of a non-zero dispersion shifted single-mode optical fiber cable

G.692 ... Optical interfaces for multichannel systems with optical amplifiers

G.703 ... Physical/electrical characteristics of hierarchical digital interfaces (includes Physical Interfaces for 2.048 Mbps E-1)

G.704 ... Synchronous frame structures used at primary and secondary hierarchical levels. (Includes Framing Specifications for 2.048 Mbps E-1)

H series recommendation - Audiovisual and multimedia systems

H.261 ... Video codec for audiovisual services at p x 64 kbit/s

H.263 ... Video coding for low bit rate communication

H.320 ... Video/audio conferencing over switched digital telephone networks

H.322 ... Video/audio/data conferencing over QoS packet switched networks

H.323 ... Video/audio/data conferencing over non QoS packet switched networks

H.324 ... Video/audio conferencing over POTS

V series recommendation - Data communication over the telephone network

V.34 ... A modem operating up to 33 600 bit/s on the switched telephone network

V.90 ... Digital modem and analogue modem pair for use on the Public Switched Telephone Network (PSTN) at data signaling rates of up to 56 000 bit/s downstream and up to 33 600 bit/s upstream

X series recommendation - Data networks and open system communication

X.21 ... Interface between Data Terminal Equipment and Data Circuit-terminating Equipment for synchronous operation on public data networks

X.21 ... bis Use on public data networks of Data Terminal Equipment (DTE) which is designed for interfacing to synchronous V-Series modems

X.25 ... Packet Switched Public Data Networks (PSPDN)

X.75 ... Packet switched signaling system between public networks providing data transmission services

X.400 ... Message Handling System (Electronic Mail)

X.500 ... Distributed electronic directories (Directory Services)

European and North American Digital Hierarchies Standard andwidth

DS-0 ... 64 Kb/s

DS-1, T-1 ... 1.544 Mb/s

DS-3, T-3 ... 44.736 Mb/s = 28 * T-1

E-1 ... 2.048 Mb/s = 30 voice channels

E-2 ... 8.448 Mb/s = 4 * E-1

E-3 ... 34.368 Mb/s = 16 * E-1

STM-1, OC-3 ... 155.52 Mb/s

STM-4, OC-12 ... 622.08 Mb/s

STM-16, OC-48 ... 2.5 Gb/s

STM-64, OC-192 ... 10 Gb/s

OC-768 ... 40 Gb/s

European/SDH standards - E-1/2/3 and STM 1/4/16/64

US/SONET standards - DS-0/1/3, T-1/3 and OC-3/12/48/192/768