600 Mbps Ethernet vs OC12 SONET

Organizations requiring fractional Gbps bandwidth have traditionally turned to OC12 SONET fiber optic service. It’s still an excellent high speed option, but you should also take a look at 600 Mbps Ethernet service. You may be surprised how the two competing connections compare.

OC-12 (622 Mbps) is the next step up from OC-3 (155 Mbps) in the SONET optical carrier hierarchy. You’ll find this service readily available in Metropolitan Area Networks (MAN) and as a dedicated Internet access connection for mid-size Internet service providers. Large organizations also have OC12 bandwidth connections, especially those with thousands of employees on-site or a need to support high bandwidth applications like medical image transmission or video production and distribution.

Not too many years ago, OC12 was considered a good choice for a carrier fiber optic core network. Now, most regional and national carriers have moved up to OC-48 at 2.5 Gbps or OC-192 at 10 Gbps. Long haul and undersea fiber lines are running at OC-768 (40 Gbps) and moving quickly to support 100 Gbps.

OC12 is most often deployed on a protected SONET ring topology, with two counter-rotating fiber rings. If something in one ring fails, either equipment or a cable break, the other ring will pick up 100% of the traffic within 50 mSec.

Note that an OC12 service is probably not the only traffic on a particular fiber strand. At 622 Mbps, OC12 comes nowhere near using available bandwidth of the fiber. SONET is designed as a Time Division Multiplexed (TDM) system that can easily add or drop common OC service levels such as OC3 and OC12. In addition to the electrically multiplexed services on the fiber, each fiber may also be optically multiplexed into a dozen or more wavelengths using Dense Wavelength Division Multiplexing (DWDM) equipment.

Carrier Ethernet over fiber is based on packet switching rather than time division multiplexing. Without the strict synchronized channels of TDM, Ethernet is far more scalable. You can order 600 Mbps Ethernet as a replacement for OC12 SONET. You can also order 500 Mbps if that is all you reasonably need. Even more importantly, you can start increasing bandwidth to 700 Mbps, 800 Mbps, 900 Mbps and 1 Gbps as business requirements dictate.

With SONET, the service provider will install a managed router as Customer Premises Equipment (CPE) that includes an interface module specifically designed for the particular OC service level you have installed. The modules for OC3, OC12, and OC48 will plug into the same port on the router, but they are not interchangeable. You need to match the particular interface to the OC service level. That becomes important when you want to implement a service change. The provider will need to roll a truck to either swap out interface modules or completely replace the CPE router.

With Carrier Ethernet, your service provider will install a managed router that can support a particular port speed. For Ethernet over Copper, this is likely 100 Mbps. For fiber, you are likely looking at 1 Gbps, although you can often get equipment rated to 10 Gbps if you expect to be upgrading to that speed level in the foreseeable future. With a 1 Gbps port, you can order whatever service level you need between, say, 100 Mbps and 1 Gbps, in 100 Mbps or smaller increments. No equipment changes are needed as you change service levels within that range. You simply call your provider and tell them to increase bandwidth and you’ll see it happen within a few days, sometimes within a few hours.

Beyond scalability, Ethernet has a cost advantage over SONET in most situations. Sometimes the cost difference is dramatic, even half the cost, for Ethernet versus SONET of the same bandwidth. The one limitation that you’ll run into is that SONET has been around a lot longer and is more available than Ethernet over Fiber. Even so, competitive carriers have been aggressively building out their IP networks and may have fiber service closer than you think.

Do you require fractional Gigabit bandwidth services? If so, it is worth your while to compare 600 Mbps Ethernet to OC12 SONET for cost and availability. The same advice is true for other service levels from T1 copper bandwidth on up to 10 Gbps and higher.

Ethernet Networks Interconnect in Arizona

Ethernet is rapidly becoming the new metropolitan and wide area network protocol of choice. Indications are that IP networks transporting standard Ethernet services will eventually supplant the century old Public Switched Telephone Network (PSTN). What’s holding that back is that many Ethernet service providers are a diverse mix of incumbent and competitive carriers with service footprints that only cover a given metro area or regional group of states. E-NNI is the technology that is linking them all together.

E-NNI is an Ethernet Network to Network Interface. Network interfacing is the cross-connection that allows different networks to exchange traffic. Each has to advertise the other’s routes so that users on either network can reach destinations on the other network as easily as they do on their own network. The result is an interconnected network with a service footprint the size of both networks combined.

An excellent example of an Ethernet to Ethernet network connection is taking place in the Phoenix, Arizona metropolitan area. Integra Telecom and Saddleback Communications are connecting their fiber networks using enterprise-class Carrier Ethernet technology so that they can easily exchange data at high transmission rates.

Saddleback Communications is an Incumbent Local Exchange Carrier (ILEC) serving the Salt River Pima-Maricopa Indian Community (SRPMIC) in the Phoenix metropolitan area, bordered by Scottsdale, Mesa, Tempe and Fountain Hills. Their core network infrastructure is comprised of multiple fiber optic rings and two Class 5 Soft Switch Central Offices. Unlike most commercial areas, 100% of the commercial developments on the Pima corridor of the SRPMIC are fed with both fiber optic and copper facility connections.

Saddleback’s business services include dedicated Ethernet Internet access, Ethernet transport services and private line services. Their BusinessLAN Ethernet Transport transparently connects business locations via native Ethernet at speeds from 5 Mbps to 1000 Mbps (GigE) with scalability in 5 Mbps or 100 Mbps over the same interface.

What Integra Telecom brings to the table is an extensive fiber optic network that serves 11 Western states, including Arizona, California, Colorado, Idaho, Minnesota, Montana, Nevada, North Dakota, Oregon, Utah and Washington. Their 5,000 mile long haul network is one of the largest in the U.S., complemented by a 3,000 route mile (160,000 fiber miles) metropolitan area network that serves 200 business communities including 1,700 fiber-fed buildings. Integra is the 4th largest Competitive Local Exchange Carrier (CLEC) in the United States.

Integra offers Metro Ethernet Forum (MEF) standardized Ethernet services that include Ethernet Private Line (EPL) and Ethernet Virtual Private Line (EVPL), as well as a layer 2 configuration for high performance Wide Area Networks (WANs). Class of Service (CoS) technology makes voice/data/video converged networks possible so that sensitive real time packets aren’t overwhelmed by less critical but more numerous data packets.

Both Saddleback and Integra operate high performance fiber optic networks and both serve a wide range of customers, including corporate clients that need highly reliable high bandwidth Ethernet services. By interconnecting their networks for Ethernet to Ethernet transport, they can offer companies on both networks the opportunity to connect as if they are working under a common “digital roof.”

Does your business need reliable high bandwidth network connection service? Carrier Ethernet now spans the nation and the world with network to network interconnections. Get competitive pricing for fiber optic Ethernet network services now. You may be surprised by what is available.




Note: Map of Arizona courtesy of Wikimedia Commons.

Fiber Optic Direct Connections To The Cloud

The Internet offers an extensive and cost effective means to access services in the cloud. What it doesn’t offer is consistent low latency, low jitter and high bandwidth performance. It was never intended to. The Internet’s claim to fame is universal connectivity. If your business processes in the cloud suffer from inconsistency or have become sluggish compared to when they were hosted onsite, you could benefit from a direct connection to the cloud.

One of the most popular cloud services is the Amazon AWS Cloud Computing Platform. Leveraging the enormous computing facilities owned by Amazon, users can pay as they go for application hosting, Web hosting, backup and storage, content delivery, enterprise IT applications and scalable database solutions. This nearly unlimited resource set can place a considerable load on typical bandwidth links used by corporations. It gets worse as you move more and more of your business processes to the cloud while leaving your employees in-place, often thousands of miles away.

The Internet works just fine for e-commerce, with thousands or millions of individual customers accessing centralized servers from their homes and offices. It’s also well suited for accessing company resources by remote workers and traveling employees. Where it starts to stumble is when you expect consistent real-time performance between the cloud and your data center or user base similar to what you get from hardwired connections in-house. The closest thing to your LAN is a high performance direct connection.

There are a couple of ways to achieve that direct connection. Amazon is currently offering direct connection to its cloud at Equinix’s Ashburn, VA and Silicon Valley, CA colocation facilities. Equinix provides colocation facilities for over 4,000 enterprises, cloud, content and financial companies through its network of colo centers. If you happen to have your equipment in their Ashburn or Silicon Valley facility, a simple cross connect is all that’s needed to hook your servers to Amazon’s.

If you are not collocated in an Equinix data center, you’ll need a WAN connection. AWS has two approved carriers that can make the connection. One is Level 3 Communications, a leading international fiber-based carrier with services to over 125 international markets. The other is AboveNet, Inc., another leading fiber optic carrier with 2.3 million metro fiber miles and connections in place to more than 450 data centers. AboveNet is offering connections to the AWS Direct Connect service at 100 Mbps, 1 Gbps and 10 Gbps.

AWS Direct Connect network service works withe the same AWS services that you would otherwise access over the Internet. These include Amazon Simple Storage Service (Amazon S3), Elastic Compute Cloud (Amazon EC2), and Amazon Virtual Private Cloud (Amazon VPC). You can use AWS Direct Connect to create multiple logical connections between your network and Amazon VPC. You can even establish multiple VPCs over this connection if you have enough bandwidth available.

AWS Direct Connect is flexible enough to scale your connectivity as your needs change. Each connection offers 1 Gbps and 10 Gbps connectivity and you can add multiple connections to increase capacity. Amazon suggests that this approach is better than using VPN hardware that frequently can’t support data transfer rates above 4 Gbps.

Have you already moved to the cloud or anticipate such a move and are concerned that your current connectivity can’t get the job done? Perhaps a direct connection can give you the performance that you need. Get pricing and bandwidth options for fiber optic cloud connections now.

Ethernet over Copper Reviving The Local Loop

There’s been a decline in utilization of local loops as of late. Most of it is the result of residential users dropping their analog phone service in favor of going totally wireless or a combination of wireless and VoIP. This has given some phone companies the idea that local loops should just be decommissioned totally. There have even been feelers to the FCC on this idea. Sadly, the idea of decommissioning copper is coming at exactly the same time as technology advances are breathing new life into the local loop.

Someday all telecom connections to the building may be fiber. That day is not today by any means. Something like a mere 12% of all business locations are currently lit for fiber optic service. Some could be but the demand isn’t there. Others might be except for the construction costs. Others still are nowhere near a fiber point of presence and won’t be in the near future.

That leaves copper as the most popular way to bring telecom services into a facility. The standardized connection is the local loop. This is a pair of small diameter plastic insulated copper wires that are twisted together to reduce noise and crosstalk pickup. One end of the copper pair terminates at a network interface device mounted on or inside the building to be served. The other end runs to the nearest central office location up to a few miles away.

Residential local loops or subscriber lines typically have two copper pair or four wires in a small diameter cable. Business locations can have dozens of individual pairs in a binder cable. Each pair can carry one analog telephone line. Two pair can provide 24 digital phone lines or a broadband data line. A dry pair has no signal at all. It is an unpowered circuit between two points. This local loop is owned by the incumbent local phone company but can be leased by competitive carriers.

What competitive carriers are doing is deploying a new bandwidth technology on the local loops that they lease. It’s called Ethernet over Copper for good reason. The last mile connection is two or more local loops connected to EoC equipment at both ends of the line. One electronics box is mounted at the central office. The other is installed in your building.

What does Ethernet over Copper offer? It’s a way to deliver higher bandwidth than you can generally get with T1 lines, another user of the local loop. T1 lines are provisioned on one or two copper pair and deliver 1.5 Mbps of bandwidth. You need an interface box called a CSU/DSU to convert that to an Ethernet protocol that will connect to your LAN. Often the CSU/DSU is a circuit card that plugs into an edge router rather than a separate box. With Ethernet over Copper, that circuit interface is unnecessary. The signal from your service provider is already in Ethernet protocol.

EoC services are available from 2 Mbps on up to 200 Mbps. The 2 and 3 Mbps levels are good replacements for T1 lines as they offer more bandwidth at about the same price. Many companies are now upgrading to 10 or 20 Mbps as more sophisticated online applications are bandwidth hungry. Also, more content from the Internet is now video which benefits from higher bandwidth connections.

The higher Ethernet over Copper bandwidths can replace DS3 services (45 Mbps) and even fiber optic lines at the OC-3 (155 Mbps) level. In addition to a considerable savings on the monthly lease cost, these bandwidths can help a company avoid fiber optic construction costs and still get the line speed they require. By the time fractional or full Gigabit bandwidth is needed, your building may be lit for fiber optic services.

EoC is strictly a business location service as of now. Cable broadband serves the residential market that hasn’t kept a local loop in service for DSL. Businesses, however, have more demand for copper wireline circuits than ever before. Plunging costs on T1 lines and EoC services are causing even small convenience stores and other retailers to rethink satellite links in favor of higher bandwidth, highly reliable and reasonably priced copper-based connections.

Could your business benefit from more bandwidth at lower prices that you’d expect? Find out what Ethernet over Copper bandwidth services are available for your location or locations and how much they cost with an immediate online quote. You may be surprised that you can easily upgrade without having to pay more.

WAN Computer Network Connections

Computer networks are often categorized as LAN, MAN and WAN. The LAN is your local network. You design and maintain it to meet your specifications. Or, you hire a contractor to do this. When you need to connect your LAN to another LAN across town or around the world, you enter the realm of the MAN and WAN. Here’s a quick look at what’s available in the way of metro and long haul connections.

The difference between the MAN and the WAN is largely a matter of degree. The MAN is a Metropolitan Area Network. As the name suggests, MAN services are limited to a particular city or city and suburbs. It’s the right connectivity for companies that have multiple offices in the same city, a main facility and a remote data center across town, a factory or warehouse in an industrial park plus some satellite offices. Hospitals and their associated medical centers are good candidates for MAN service, especially higher bandwidths for medical image transmission.

A difference between the LAN and MAN is that you own your LAN and buy services from the MAN. A Metropolitan Area Network is typically privately owned and offers services to numerous clients. The network operator manages bandwidth and security to ensure that every customer feels like they have the network all to themselves.

Most MANs are organized as SONET rings that encircle the metropolitan area. A SONET ring is actually two fibers with the same traffic running in opposite directions. One is the primary service and the other is the backup. This is done to ensure reliability. If one fiber or its associated equipment experiences a failure, service is automatically switched to the other fiber in the ring within 50 milliseconds. SONET has designated service levels at OC-3 (155 Mbps), OC-12 (622 Mbps), OC-24 (1.24 Gbps), OC-48 (2.49 Gbps), OC-192 (9.95 Gbps) or OC-768 (39.8 Gbps). You may also be able to get DS3 service multiplexed onto an OC-3 service for transport across town.

Newer metro networks may offer Carrier Ethernet instead of SONET. Ethernet is highly scalable from as low as 1 Mbps on up through 10 Gigabit Ethernet. Popular MAN service levels are 10 Mbps, 100 Mbps, and 1,000 Mbps or GigE. An advantage of Ethernet is that changes in service level can be accomplished faster than with traditional SONET. You can also get layer 2 Ethernet connectivity, such as E-Line and E-LAN, to connect your LANs at other locations into one large bridged network. The MAN itself may have an IP core or be running as Ethernet over SONET to use existing facilities.

Think of the WAN as a very large MAN and you probably won’t be far off. WAN networks may also have SONET rings or IP core networks. They are certainly fiber optic based and often with so many strands that you can lease wavelengths or even dark fibers. Many WAN networks are structured as MPLS or Multi-Protocol Label Switching networks for flexibility. A MPLS network has inherent security due to a propriety routing technology that is not IP. MPLS networks can transport any protocol by encapsulating the packets within the switching labels.

WAN networks may have a regional, national or international footprint. Some are designed with low latency paths that are especially important for financial trading. WAN networks may also peer with other networks of similar size to share traffic. The newest wrinkle is the Ethernet Exchange that exchanges Ethernet traffic without having to first convert it to SONET.

How do you connect to a WAN network? Often you can pick the connection that is both readily available and lowest in cost for your particular application. These range from T1 lines to Ethernet over Copper, Ethernet over Fiber, and SONET last mile connections. In some cases, fixed wireless access is also available. The mix of copper, fiber and wireless access should accommodate just about any need.

Do you have a need to connect your LAN to another facility across town, or to locations in other states or countries? Compare prices and availability of various MAN and WAN connectivity options to meet your requirements.

Managed Hosting in Clouds and Colos

When it comes to hosting, you have all sorts of options. Most individuals and smaller companies opt for shared hosting. It provides decent performance at a rock bottom price. Once you get too big for shared Web hosting, you’ve got a decision to make. Do you do it yourself or opt for a managed solution?

There was a time when you needed the savvy to run your own web server to even get a site up and running. Now that Linux and Windows hosting has become so common and standardized, there are lots of places to get hosted. Even larger companies that insisted on maintaining control by buying their own servers and rack mounting them in their own temperature controlled data centers are taking a second look at colocation and clouds. Why? It’s mostly about cost but also about resources.

One of the big resource bottlenecks today is bandwidth. Certainly, carriers have kept up with offerings at GigE, 10 GigE, OC-768, wavelengths and dark fiber. What they haven’t done is provide universal access. While competitive fiber optic networks are expanding their service footprints every day, the majority of business locations still aren’t lit and aren’t likely to be in the near future. Ethernet over Copper bridges the gap for some. Speeds are up to 200 Mbps now. EoC is distance limited, however, so that your best chance for service is in a downtown business district.

Move to a colocation facility or cloud service, however, and your bandwidth issues may be over. They may not be if you need a high bandwidth pipe between your facility and the cloud. But if most of your bandwidth demand is coming from Internet users rather than in-house users, colos and clouds look pretty attractive. Cloud providers locate with the same facility at major carriers to ensure themselves of almost unlimited bandwidth. You can do the same thing by packing up your high bandwidth demand servers and shipping them to a colo facility. The best deals are where multiple carriers have established points of presence and are willing to bid for your business.

Another attraction of colocation is jettisoning the capital investment and operating costs associated with running your own data center. The colo has high security, backup power, environmental control and a tech support staff available 24/7. You need to provide the same things. Economy of scale favors the colocation company with its much larger facility and lots of customers to amortize the cost.

Smaller companies may find that they can’t afford an around the clock tech staff nor the investment required to build or expand an in-house data center. A move to a nearby colo center can get them the facilities they need for a monthly fee. But why stop there? Perhaps it makes even more economic sense to forget about having your own hardware at all. Why not pay as you go on everything?

This is the appeal of everything-as-a-service. Hedge your bets by renting rather than buying. You can do that at many colocation centers now. They’ll put a server in the rack for you and keep it maintained. It’s just like having your own hardware except that when you don’t need it anymore, you just walk away. Need a bigger server? Don’t buy one. Simply upgrade your colo service.

The cloud does the colo one better. The cloud philosophy is “why commit to any particular hardware at all?” Why, indeed? In the cloud all services are virtualized. You don’t need to know or care what they’ve mounted in the racks. What you are concerned about is how many instances of virtualized servers you need at the moment. If you find that your demands fluctuate, you can increase or decrease the number of servers or amount of storage almost instantly. The well of resources to tap is nearly unlimited.

The problem now is how to sort out the options. Shared hosting is nearly a commodity these days. Get competitive quotes for colocation and cloud hosting services for your IT operations and then compare with what it costs you to provide the same value in-house.

EoC and EoTDM Complement Fiber

Traditional telecom services have large bandwidth gaps in service levels, especially when you compare what is available for smaller businesses and what the large corporations install. You may be slogging along at T1 lines rates of 1.5 Mbps, while the big corporate campus across the street has 100x to 1000x the bandwidth. It’s the difference between copper and fiber, or at least it used to be that way.

Newer technologies, namely Ethernet over Copper (EoC) and Ethernet over TDM (EoTDM), have expanded the range of bandwidths supported on already installed copper infrastructure. It’s now a smooth scale of steps between T1 and Gigabit or higher speeds.

Integra Telecom is one competitive carrier who has embraced the concept of augmenting their fiber optic network with Ethernet copper-based solutions. Integra is a major telecom services provider in eleven states, primarily in the Western US. Their 10 Gbps IP fiber backbone connects major cities throughout these states and also to Chicago, New York, Dallas and Ashburn, Virginia. It totals 5,000 miles of long haul and 3,000 miles of metro fiber.

From there, Integra is aggressively lighting new buildings for fiber service. Just during the first half of 2011, they had expanded their lit buildings by 20 percent or 300 new buildings. Even the most casual observer can see that fiber-based bandwidth is the future. Business automation, cloud connections and video transport are gobbling up Mbps and Gbps about as fast as they are deployed. But what do you do in economically constrained times to provide higher bandwidth levels to every business that needs them?

The answer is a two-pronged approach of complementing fiber network resources with copper-based last mile connections. Traditionally, this has meant T1 lines with DSL as a low end option. T1 provides excellent availability and rock solid bandwidth, albeit at 1.5 Mbps. Many businesses get around this limitation by installing multiple T1 lines bonded together to double or triple the bandwidth. This is a solid technical approach, but the upper limit is typically around 10 or 12 Mbps and the cost is a simple multiple of T1 cost times number of bonded lines.

Another hitch is that most competitive metro and wide area networks are now IP based. Previously the public switched telephone network was dominant using telco standards like TDM for digital transmission. Businesses today have IP networks in-house and often wish to connect them to other IP networks through Ethernet services such as E-Line and E-LAN. It would be ideal to keep everything on IP networks.

Integra Telecom recognizes this and bases their fiber/copper strategy on IP standards. The core network is already IP for metro and long haul connections. Where copper connections are needed, they are made using Ethernet over Copper or Ethernet over TDM.

The copper involved is the very same twisted pair copper that is already installed into nearly every business for multi-line telephone service and T1 data lines. That pretty much guarantees that businesses in the service footprint can get connected to the network one way or the other. Ethernet over Copper uses specialized equipment and multiple copper loops to deliver Ethernet bandwidth from the fiber network to the business LAN. Ethernet over TDM is similar, but uses T1 line technology to transport the Ethernet signal.

Why EoTDM? One tradeoff with EoC is that you can get a very wide range of bandwidths but over a limited distance. The closer you are to the central office that connects to your copper telco lines, the higher the EoC speeds available. EoTDM is not distance limited. T1 technology was designed for extended distances experienced by some customers who are located in smaller or more rural population areas. Integra has installed EoC equipment in 120 COs and EoTDM in nearly 300 COs throughout their Western service area.

Are you ready to move up to higher bandwidth levels, but stymied because you’ve been told there’s no fiber available? That may be old information, as carriers are aggressively building out their fiber assets. Even if fiber is not an option, you may be able to get bandwidth from 2 or 3 Mbps to over 100 Mbps on copper based connections to the fiber backbone. Check prices and availability of copper and fiber network services for your particular business locations.

Telecom Providers Build Their Own Clouds

Just as cloud companies are beginning to mature as businesses, Internet service providers and other carriers have decided to build out their own cloud offerings. Now, how does a harried IT manager or small business owner sort through all the various opportunities?

It’s important to remember that we are in the early days of the cloud. Like all new technologies, this one has a maturity or learning curve. We are on the early part of the upward slope. What that means is a lot of activity from a lot of players. You can expect new companies with new services to be popping up all the time. It also makes sense that current players in the computer and networking space will want to corral as much of this business for themselves as possible. If not, they may be justifiably concerned that they could be relegated to sidelines.

Windstream, a major competitive carrier for T1 lines, MPLS networks, dedicated Ethernet Internet and enterprise VoIP, saw the handwriting on the wall when they acquired Hosted Solutions last year. Adding the assets of Hosted Solutions to Windstream’s existing data centers, has given them the critical mass to get into cloud services for existing customers. Such services include such things as cloud storage, Infrastructure as a Service and private, public, and hybrid clouds. Other carriers, such as Level 3 Communications, also have cloud services available.

Most carriers have large data centers for their own use. They often leverage these assets by offering colocation services within their secure and highly reliable facilities. Colocation is something of a forerunner of the cloud. The provider offers racks and cages where you can move your servers and network appliances. They provide the electricity with backup, environmental control and security. Another big draw of colocation is the proximity to large amounts of bandwidth. Many companies either can’t get or can’t afford the cost of constructing fiber optic connections to their own facilities. At a colocation center, the carrier is right down the hall and a mere cross-connect away. It’s the best deal on bandwidth you can get.

More recently, colocation centers have begun to offer contracted technical support and even leased servers for those who don’t want to buy their own. In essence, the colo becomes your data center and you don’t need the capital expense, operational expense or staffing to run your own. That being the case, what’s different about cloud services?

The primary difference between colocation and the cloud likes in both outsourcing and virtualization. The cloud infrastructure consists of massive computing power and storage, all virtualized so that it can be sliced and diced as users require. While in the cloud, you are unaware that you are not the only one using the facilities. The same bank of servers that run your applications can be running dozens or hundreds of others simultaneously. The magic of virtualization creates the illusion that you have one or more physical servers all to yourself.

A good cloud is much more than that. Not only do you rent rather than buy, but you rent by the minute or hour times the number of servers you are using. You can add or subtract virtual servers at will and only pay for the ones you are reserving. The same is true for storage. You don’t worry about buying a new disk when you fill up the one you have. You simply increase or decrease storage as needed and pay by the byte.

The ability to increase and decrease resources almost instantly is a feature unique to the cloud. This scalability is highly desirable for companies with varying loads or ones that are rapidly growing. There is no need to be constantly buying and upgrading equipment when you can simply log into your cloud account and add resources at will.

What carriers bring to the table is one stop shopping. They already provide last mile access, multi-site connectivity, and converged voice, video and data networks. By adding cloud services, you have one bill to pay and a single point of contact for resolving issues such as latency or availability. Independent cloud service providers will need to be on their toes to stay ahead of the carriers, by offering more advanced services and lower pricing to stay in the game long term.

Are you ready for the cloud? The range of services and competitive pricing makes cloud computing and storage, colocation and managed services more cost effective than they’ve ever been. Inquire about availability and pricing for the networking and computing services you need for your particular applications.

Cloud Service Providers Colocate With Customers

There’s a new migration on. No, not the migration from local data centers to the cloud. It’s a migration of cloud service providers to colocate with their customers.

Why? It’s the rude awakening that companies suffer when they find that applications running way out there don’t have the same zip as the ones running on servers down the hall. That can be fixed, of course. It requires engineering your connections to the cloud so that they have sufficient amounts of bandwidth and very low latency, jitter and packet loss. But why go that that expense when you can just cozy up to your service provider with little more than a long patch cord connecting you?

That’s what Telx thinks. They’re throwing open their doors and working to lure cloud service providers to take up residence in one or more of their high performance colocation centers. It’s an acknowledgement that the speed of computing has gotten to the point where interconnections are the weak link. The venerable T1 line or DS3 connection that worked so well when all you needed to do was surf the Web, exchange email or generate sales leads quickly chokes when you start adding virtual servers by the dozens or hundreds.

There are a couple of big crunches that strain the cloud service model. One is business processes that are now automated and support hundreds or thousands of employees. The other is high performance e-commerce, with enormous catalogs and thousands of simultaneous shoppers who have little patience for sluggish servers and none at all for errors in their shopping experience.

The cloud is becoming a victim of its own success. Cloud service providers have mastered the technology of being able to ramp up or down the number of servers online in a matter of seconds. Storage is a bottomless well that data fills as needed. Public-facing bandwidth is easily handled with multiple Gigabit and 10 Gigabit Internet connections with enough margin to serve all customers. While cloud companies have gotten used to this scale of resources, most businesses have not. They don’t have the private line connectivity to take full advantage of near-infinite, near-instantaneous computing resources.

More savvy companies have mastered the connectivity issues, but who’s going to turn down the opportunity to get more performance for less money? The notion of being on the same floor of the same building with your service provider and ditching the telecom line in favor of a local fiber or wire connection has a lot of appeal. The issues of bandwidth, latency, jitter, and packet loss simply disappear.

A lot of medium and larger companies already have equipment in colocation centers, such as Telx. The economic tradeoff between running your own environmentally controlled facilities with backup power, fire suppression and round the clock monitoring and simply renting space in a larger facility with its economies of scale makes colocation attractive. On top of that is the fact that you have easier access to competitive carrier services with more attractive pricing than you can get locally. Now, add to these advantages the opportunity to connect to a cloud service provider in-house and you are looking at really attractive cost vs performance figures.

The new Telx Cloud Connectivity Centers with cloud-optimized infrastructure to support cloud service providers in each of their 15 data centers is an idea who’s time has come. Enterprise users and cloud companies are like opposite magnetic poles trying to get as close to each other as possible. Colocation is the obvious answer.

Can your business benefit from colocation as a service provider or service user? Get pricing and technical specs for the IT resources you need and compare with what you are doing now.



Note: Photo of clouds and building courtesy of Wikimedia Commons.

Fiber Optic Bandwidth For Remote Backup

Most companies have or should have a plan for protecting their valuable data against disaster. Simple disk and tape backups may protect against a crashed PC or server, but offer little or no protection when you get hit with fire, flood, tornado, hurricane or earthquake. For this reason, offsite is the way to go for data backup.

The question become how far away is safe and what do you need to efficiently transfer data. The data center across town will save you if your building catches fire, but won’t do much when an F5 tornado levels the entire town or the levee breaks and everything submerges. You need to get to a different geographical area where the threats are different than where you are now. Tornados aren’t likely to hit Denver and Biloxi at the same time. A big earthquake may shake San Francisco, but it’s not likely to be shaking Chicago at the same time.

Corporations with regional offices are well situated to back each other up. You may want to consider having each office back up data for a couple of other offices hundreds or thousands of miles away. It’s a lot of data flowing, but if one office gets taken out, you’ve got the data in at least two other places.

Having geographically diverse locations back up each other’s data fits well with the network topology you need anyway. Certainly, you want all of your locations to be able to communicate. Some companies do this through a private star network centered on their headquarters location. Others use a mesh network based on MPLS with each location connecting to the nearest POP in their city. That’s more robust because there is no single point of failure like you have with HQ coordinating all the traffic.

What you need for data backup is bandwidth and lots of it. With dedicated line services you are paying for the full capacity even though the lines mostly sit idle after hours. Those are the traditional times that IT departments do data backups anyway. You can load your lines to the max to transfer data files without affecting other business operations if you do this in the wee hours.

Other companies like the idea of data mirroring. This is a real time process that keeps two or more copies of a data set completely synchronized at different locations. The advantage is that you can recover almost instantly when you lose a facility. The tradeoff is that you’ll need much higher bandwidth levels than for backup file transfers that can take their time overnight.

What type of bandwidth is available to support your backup or mirroring needs? Most likely you’ll want fiber optic connections, although smaller organizations may get by with 100 Mbps Ethernet over Copper where it is available. Otherwise you’re looking at traditional SONET services that include OC-3 at 155 Mbps, OC-12 at 622 Mbps, OC-48 at 2.49 Gbps or OC-1921 at 9.95 Gbps. A good alternative is Carrier Ethernet service that is highly scalable and includes popular speeds such as 100 Mbps Fast Ethernet, 1000 Mbps Gigabit Ethernet (GigE) and 10,000 Mbps 10 Gigabit Ethernet ( 10 GigE).

Both SONET and Ethernet are widely available for long haul connections through competitive fiber optic carriers, plus last mile connections to high bandwidth MPLS networks with a regional or nationwide service footprint. Which topology you pick depends on how many locations you want to interconnect and how the pricing works out. Ethernet tends to be less expensive than SONET in most situations and MPLS networks have the advantage of offering more bandwidth at lower costs than most dedicated lines.

Fiber optic connections are also valuable if you are getting your storage in the cloud, as many companies are now doing. The advantage of cloud storage is that it gives you the geographical diversity you require, the security of a major data center and nearly unlimited resources as your data storage needs change. You’ll also avoid the capital expense and staffing required to run your own data centers.

Are you in need of robust high bandwidth connections to support your data backup and mirroring needs? If so, get pricing and availability of SONET and Ethernet fiber optic bandwidth services now. You may also wish to explore cloud storage options as another approach to disaster protection and recovery.



Note: Data center photo courtesy of Wikimedia Commons.

Fiber Buildings Offer High Bandwidth

Are you looking to increase your bandwidth connections from the low to medium speeds you can get delivered over twisted pair copper or fixed wireless, on up to the high speeds available only over fiber optic cabling? Would you like to do that while avoiding the often prohibitive construction cost of bringing in new fiber service? Then what you need is a fiber building.

In this case, a fiber building isn’t one that’s made out of carbon fiber construction or anything exotic like that. Fiber doesn’t refer to the construction materials of the structure itself. It refers to the type of telecom services already installed. A fiber building is one that already has fiber optic service brought in and operating. That’s why you often hear them referred to as fiber lit buildings. Lit means that there is laser light shining down those fiber strands and carrying digital payloads.

So, how do you find a fiber building? Peeking in windows after dark won’t work and will probably get you in trouble. You could start calling all the office and industrial buildings in the area and mark the lit building locations on a map. Or you can take the easy approach and use an automated online tool that will do all that legwork for you. It’s called the “Lit Building Locator.” You simply enter the street address of the building you are in now and you’ll get a map of lit buildings in the area and how far they are from you.

The most popular fiber optic bandwidth service being requested now is Metro or Carrier Ethernet. The beauty of Ethernet is that it is the same protocol that you are already running on your corporate LAN. That makes interfacing very easy. Fiber optic Ethernet services include Fast Ethernet at 100 Mbps, Gigabit Ethernet (GigE) at 1,000 Mbps and 10 Gigabit Ethernet (GigE) at 10,000 Mbps. In some cases 40 Gigabit Ethernet is also available for the most demanding applications.

You should be aware that there is another type of fiber optic service that is also installed in “lit” buildings. This is the traditional telecom service called SONET for Synchronous Optical NETwork. Bandwidth services include OC-3 at 155 Mbps, OC-12 at 622 Mbps, OC-24 at 1.2 Gbps and OC-48 at 2.4 Mbps. Higher speeds include OC-192 at 9.95 Mbps that is compatible with 10 GigE WAN-PHY and OC-768 running at 40 Gbps.

Which fiber service should you choose? There may or may not be a choice. Once a fiber service is brought into a building, competing services generally choose other buildings to light up. Any particular building may be lit for SONET or Ethernet. There is a service called Ethernet over SONET, including 10 GigE WAN-PHY, that can give you the protocol you want even if your building is set up to legacy telco standards. Another reason that customers are asking for Ethernet is that it tends to be less expensive per Mbps than SONET services, where available.

Are you looking for more bandwidth? Why not check for fiber buildings and see if yours or one nearby is lit? If close, you may be able to get a carrier to make the necessary installation with no or discounted construction costs if you will be ordering a high bandwidth level. Perhaps you can find other tenants and pool your request to make the installation even more attractive.



Note: Photo courtesy of Diego Silvestre on Wikimedia Commons.

Telx Cloud Exchange Reduces Service Lag Time

The idea behind cloud services is that you can shut down your local data center and rent everything you need in the way of infrastructure, platform and software from a service provider somewhere out there. You’ll save a ton of money and the users will never know the difference. Except some of them do notice a difference. Things can bog down in the cloud.

How can that be? One premise of cloud computing is that there is a near infinite well of resources to draw from. You need more processing, you bring it online almost instantly. You need more disk storage, you take it. After all, there is plenty to go around. So how can responses that were snappy when the data center was in the basement become sluggish when the same facilities are in the cloud?

It all comes down to connectivity. Everybody knows that electrical signals move at the speed of light, right? You might guess that’s so fast that it shouldn’t make any difference if the wire is a hundred feet long or a thousand miles. At the speed of light a human can’t possibly detect the travel time of electrical impulses over wires and fiber optic cables. That’s right, isn’t it?

It sure sounds nice in theory, but in practice the speed of light isn’t infinite and the speed of communication signals never gets near light speeds anyway. Remember that the speed of light so often quoted is in a vacuum. That’s 186,000 miles per second when your laser beam is shooting through space. On terrestrial circuits, 186,000 miles per second translates to 186 miles per millisecond or 10 milliseconds for 1,860 miles. Those signals can’t even go that fast because any medium slows them down. You’ll be lucky to go 2/3 as fast, or a millisecond for every 124 miles.

Are we forgetting something? You bet. There’s no such thing as communicating over one long strand of pure wire or fiber. There’s circuitry at both ends and amplifiers, regenerators, add-drop multiplexers and other equipment in-between. Those will add milliseconds or tens of milliseconds more.

That’s still nothing compared to what happens when packets are routed on the Internet. They get from point A to point B alright. But they seldom go in a straight line. They go from router to router to router to router and eventually to the destination. There’s no guarantee the next packet will take the same route as the last one. There’s also no guarantee that the packet will even get there intact. Oh, one is missing? TCP/IP will resend it and all will be well. The file being transferred will certainly be intact at the other end, but how long did it take to replace all the lost packets and wait for traffic jams congesting certain nodes?

Cloud providers and companies sensitive to lag time, also known as latency, are taking a close look at colocation to have the shortest and most direct communications paths possible. A step beyond even standard colocation facilities is the new cloud exchange service from Telx. It’s branded cloudXchange and it may be the future of data centers.

Telx’s breakthrough comes in inviting cloud service providers to move in with them, literally. A service provider can locate their infrastructure in one or all 15 Telx facilities. What they gain is access to a wealth of carriers who have created point of presence within the Telx facilities and major corporations, content delivery networks and others who are just down the hall in the same building. For long haul connections, Telx has access to low latency fiber routes between data centers and to worldwide destinations.

This may be where we’re all headed. Instead of every company having its own server racks connected directly to the corporate LAN, most infrastructure will be outsourced to a cloud service provider or collocated in the same building to form a hybrid cloud. User connectivity will be over high speed dedicated lines, perhaps just to the nearest colo facility where service providers will have a portion of their infrastructure. A separate Internet access path will be available to browse the Web, share email with outsiders, and connect with consumers.

Are you a user or provider of cloud services who is unhappy with their networking connections? Perhaps you can benefit from an upgrade to higher bandwidth, lower latency connectivity to get rid of the lags that are plaguing your business processes.



Note: Photo of data center servers courtesy of WikimediaCommons

Gigabit Ethernet vs SONET

With the expansion of video content and moves to cloud services, business bandwidth now easily pushes the Gigabit per second level. This certainly means a move from copper to fiber optic services. But did you know that there are two types of fiber optic bandwidth to choose from? They are SONET and Gigabit Ethernet.

SONET is the original fiber protocol developed by the telephone companies to transport large numbers of simultaneous telephone calls between switching centers and internationally. SONET stands for Synchronous Optical NETwork. There is an international version of this protocol called SDH for Synchronous Digital Hierarchy. Together they are often referred to a SONET/SDH.

SONET’s origins go back to the initial development of digitized telephone conversations and the transmission of these a groups using trunks known as T1 lines. The larger T3 lines can carry the equivalent of 28 T1 lines or a data bandwidth of 45 Mbps. There are no faster copper telco standards. The next step up is OC-3 at 155 Mbps. That’s enough to transport 3 DS3 circuits and, indeed, the protocol is designed to easily multiplex and demultiplex lower level services. When you order a DS3 today, it will be transported most if not all of the way multiplexed into a OC service. A device called an ADM or Add-Drop Multiplexer gets the service on and off the optical network.

You’ve probably guessed that OC stands for Optical Carrier. There are a number of optical carrier levels that include OC-3 at 155 Mbps, OC-12 at 622 Mbps, OC-24 at 1.24 Gbps, OC-48 at 2.49 Gbps, OC-192 at 9.95 Gbps and OC-768 at 39.8 Gbps. When you hear about 10 Gbps SONET, it is an OC-192 service. The 40 Gbps that many carriers use for long haul transmission is OC-768.

What about intermediate rates? Not all levels between 1 and 768 are defined. Others, such as OC-9, OC-18, OC-36 and OC-96 are standardized but not generally offered in service.

SONET/SDH is set up in ring arrangements with two separate fiber strands for each ring. They are redundant circuits, so if one fails the other can pick up the load within 50 mSec.

Ethernet is a completely different protocol developed by the computer industry rather than the telephone companies. It is based on packets rather than telephone channels. Most LAN connected devices have a NIC or Network Interface Card that supports 10/100/1000 Ethernet. That means the device can run at 10 Mbps, 100 Mbps or 1,000 Mbps (1 Gbps). Corporate LANs may now run at 1 Gbps with some parts of the network using 10 Gbps or 10 GigE.

The equivalent of Ethernet in the LAN for local networks is Ethernet WAN or Wide Area Network. Switched Ethernet used for WAN applications is also called Carrier Ethernet. If the service is local to a particular city or metro area, it is often called Metro Ethernet or Metropolitan Ethernet and the network is referred to as a MAN or Metro Area Network.

Like telco services, Ethernet may be delivered over copper or fiber. Twisted pair copper telephone line can be used up to 100 Mbps. That speed is also available over coaxial copper cable through Cable systems. Beyond that, fiber optic connections can deliver 1 Gbps or 10 Gbps for business applications.

Ethernet is designed to be far more scalable than SONET. Instead of ordering a particular service level, such as OC-24, you order an Ethernet port capable of handling a maximum speed, say 1 Gbps. With the port in place, you can pay for Gigabit Ethernet if you need that much, or you can order 250 Mbps or 500 Mbps Ethernet and upgrade quickly and easily when the need arises.

Which fiber optic service should you order? In general, Ethernet is less expensive per Mbps than SONET. Sometimes the price savings can be dramatic. The one sticking point is that you need to have the service installed in your building. You may have one or the other services now or nothing at all. The cost of construction to bring in the right fiber from the right carrier can outweigh the monthly lease costs. You should know, though, that Ethernet can often be carried on SONET and that competitive carriers are hungry to get new buildings on their networks. These factors can affect the availability and pricing of fiber optic services to your building.

Fiber optic build-outs are progressing rapidly and the demand for high speed bandwidth increases. Even if you went out for bids a year ago, the situation may have changed for the better since then. The best thing to do is get new prices and availability for fiber optic bandwidth services for your particular business location.



Note: Photo of data center servers courtesy of WikimediaCommons

From CAPEX to OPEX In The Cloud

There’s a new cost model for Information Technology. It’s a move from high CAPital EXpense or CAPEX to OPerational EXpense or OPEX. What’s driving this change? It has to do with the uncertainty of the business climate and the availability of cloud computing services.

What’s wrong with the way we’ve always done it? The legacy practice of building on-site environmentally controlled data centers and stocking them with server racks, arrays of disk storage and network appliances works very well if you can come up with capital to implement this plan.

There are also operational costs in the form of electricity, bandwidth and staffing, not to mention the near-constant process of patching and upgrading software. Those software packages aren’t cheap, either.

Traditional Information Technology is capital intensive. That hasn’t been a problem until recently because it was accepted business practice and everybody did it. Then along came the “Great Recession” and everything changed. We’ve had a lot less certainty about the economy the last few years. Most companies have responded by hunkering down and minimizing cost of all types. For those that see opportunity and want to expand, there’s a dearth of available financing.

Any wonder why cloud services are multiplying before our eyes? The business climate provides the impetus and the cloud provides the solution. It’s a combination that’s turning a trend into a stampede.

What’s so great about moving out there to the cloud? For many companies, it’s a combination of agility and expense versus commitment and capital. There’s nothing going on in the cloud that you can’t do yourself. What cloud service providers offer is a way to offload, or outsource, the infrastructure, platform, software and staffing that isn’t proprietary to your business.

For smaller companies, buying cloud services gives them a robustness of computing that they’ve never had before. If you are too small to have a dedicated IT staff on duty around the clock, you’re pretty much flying by the seat of your pants. You expect that whatever is working today will still be working tomorrow. When a part of the network or process goes down, you troubleshoot as best you can or get on the phone to a consultant or the vendors. It’s not unusual to have long periods of unplanned downtime.

Not so if you get your computing and perhaps your telephone service from a reputable cloud provider. The core business of the cloud services company is to provide a limited number of services and execute them perfectly. They have the equipment, the redundancy and the staff to make sure things are always up and running. It’s what keeps their many customers buying from them. You may not need an extensive array of servers, storage, software or bandwidth, but you’ll benefit from using the same infrastructure as those who do.

For the larger organization, high performance is a basic expectation. What the cloud has to offer the enterprise is a way to increase and decrease infrastructure resources on demand. You need more virtualized servers? You’ve got ‘em. You need more storage? There’s plenty in the pool to draw from. More visitors creating a demand for more bandwidth? No problem. The fiber optic lines are already in place with more bandwidth than any one company can use.

You simply pay for what you need when you need it. That gives you a tremendous flexibility to pursue new business opportunities with the knowledge that if they don’t pan out, you are not committed to a huge computing investment that must be paid off regardless.

Could your small, medium or large enterprise benefit from moving from a capital intensive to a pay as you go expense model for information technology services? Find out by comparing prices and available services from cloud service providers. Then get started at the level that makes the most sense for your organization.

A Myriad of Business Broadband Options

Business class broadband comes in more flavors that ever before. Your options include DSL, 3G & 4G Wireless, Cable broadband, Satellite, T1 lines, and Ethernet over Copper. Those are just for small business locations. Fiber optic services expand the menu for medium and large scale business users. What is it that distinguishes these broadband options and how do you choose?

You should know that there are two different types of Internet broadband access regardless of the medium used. The medium might be wireless, twisted pair copper, coaxial cable, or fiber optic strand. That’s just the physical layer. What makes even more difference is whether you are on a dedicated or shared broadband connection.

Your lowest cost options fall into the shared bandwidth category. These include DSL, Cable, 3G & 4G Wireless and Satellite. What these all have in common is that the bandwidth is shared among multiple users. A tip-off to shared bandwidth is a statement that your service speed is “up to” so many Mbps. The upper limit, be it 2 or 20 Mbps, is the maximum you’ll experience. How much you get in practice depends on how many other people are trying to use the connection at the same time.

Shared bandwidth services don’t have to be asymmetrical, but they often are. What that means is that the upload and download speed differ, often by a large amount. You might get 5 Mbps in the download direction, but only 1 Mbps in the upload direction. Why is it done this way? It’s because these services are most often used for email and Web access. When using a Web browser, you typically type in an address and then download a page of text and graphics. The upload to request the page is a very few packets. The download of everything from the page is a much larger collection of packets.

The same is true of accessing audio and video content or downloading software. It’s primarily a one-way transmission with the speed of the download much more important than the speed of the uploaded requests.

Another characteristic of shared bandwidth services is that they are promoted as “information” rather than “telecom” services. By using the information category, these services avoid regulatory constraints. What you get for your money in the way of bandwidth levels, availability, time to repair, latency, jitter, packet loss is completely up to the service provider.

Dedicated Internet Access or DIA services allocate the full bandwidth of the connection for your usage. A T1 line, for instance, has a speed of 1.5 Mbps at all times. If you aren’t using it, the line simply idles until needed. You can elect to access the Internet via your T1 line as needed, or you can keep the line busy uploading and downloading data files 100% of the time. There’s no such thing as user limits. You’ve got a pipe with a capacity of 1.5 Mbps and you can transport as many packets through it as it will carry.

T1 lines are symmetrical, also called full duplex. The bandwidth is the same in both the upload and download directions. In this case, it’s 1.5 Mbps. Ethernet over Copper, a newer competing service to T1 lines, is similar but offers more bandwidth options. You can often get 2, 3, 5, 10 and even 20 Mbps EoC service. It’s often stated in the form 10 x 10 Mbps Ethernet to show that you get 10 Mbps in the upload and 10 Mbps in the download direction.

T1 and Ethernet services are not information services. They are network telecom services, often with SLAs or Service Level Agreements. These agreements spell out what you can expect in the way of availability, time to repair, bandwidth, jitter and latency. Because these services are more costly to engineer and provide, they have higher pricing than shared bandwidth information services.

So does that mean that you always need a T1 line or Ethernet connection? Not necessarily. If your primary reason to have an Internet connection is to access information on the Web or maintain a Website and email on a remote hosting service, you may find that DSL, Cable, Satellite or Wireless service works just fine and costs half as much. Satellite and Wireless are especially useful for remote locations that aren’t within the service areas of DSL or Cable. You can use 3G and 4G cellular wireless for point of sale terminals, even at temporary locations.

On the other hand, if you want to run your own servers in-house or connect multiple business locations with a converged voice and data network, you’ll probably find that you absolutely need the more robust bandwidth services. The same is true if you are providing rather than consuming content or can’t stand having employee productivity impacted by bandwidth that varies up and down throughout the day. Would a multi-day outage put you out of business? It probably makes sense to either have a backup bandwidth service or pay up for a Service Level Agreement.

What Internet access options make the most sense for your particular business situation? Why not have a bandwidth expert call you with prices and availability for a wide range of bandwidth solutions? Then you can weigh the costs and benefits and make an better informed decision.

100 Mbps Colocation Center Bandwidth

Companies often look to colocation as a way to avoid the capital and operating expense of running their own data centers. There is also another important incentive to go the colocation route. It’s high bandwidth availability at lower costs.

A colocation center, or colo, is a facility that serves the data center needs of many businesses. Each server farm requires the servers themselves, networking switches, routers and appliances, equipment racks, rack wiring, electrical power, cooling air, and connectivity. Plus you’ll need physical security, fire suppression and a staff to monitor operations and take corrective action 24/7. You can provide all of this in-house or get it at a colocation center.

What makes a colo center cost effective is economy of scale. Instead of each customer having to maintain their own backup generators and fire suppression systems that are rarely used, the colocation facility has much larger units that serve all customers. The costs don’t scale linearly. The incremental cost of adding another user to the facility is pretty low and the facility cost spread over all users is less than what you can do on your own.

The somewhat hidden secret of colocation centers is that they have bandwidth available that you may not be able to afford on your own. A common connection is 100 Mbps Fast Ethernet. You know that it’s no great shakes to connect a 100 Mbps Ethernet node to your own LAN. Getting a 100 Mbps connection beyond your property line may be anywhere from expensive to impossible. At 100 Mbps, you are beyond the capability of twisted pair services such as EoC or T1/E1 lines. You’ll need fiber optic cabling brought into your building and lit for SONET/SDH or Metro Ethernet service.

Why is there such a difference in price between 100 Mbps bandwidth provided in a colocation center and the same level of service delivered to your facility? One big reason is construction costs. Downtown in a major metropolitan area there may be many fiber facilities in your office building or nearby. If you order a significant amount of bandwidth you might not even be charged construction costs. But move a bit off the beaten fiber path and those charges go up dramatically. If you are located in a smaller town, industrial park or rural area you can be looking at tens of thousands of dollars or more to trench fiber optic cable from the nearest carrier point of presence.

The other reason for bargain prices at colocation facilities is competition. Once again, depending on population density and popularity of high bandwidth network services, you might find yourself with only one carrier serving your location. Your negotiating ability is pretty limited without competition. Colocation centers, however, are carrier magnets. Service providers know that there are dozens or hundreds of customers in a single building who all need their services. It’s no wonder they all set up points of presence within the colo. Often, there is a “meet me room” where carriers and customers are connected.

The combination of having several or more eager carriers vying for your business combined with the trivial construction costs of running a drop to your equipment racks is what makes it easy and relatively inexpensive to get all the bandwidth you want in a colocation center. Can you say the same for where you are now? There’s an easy way to find out. Simply request bandwidth availability and pricing for both your location and a nearby colocation facility.



Note: Photo of server rack wiring courtesy of Guillaume Paumier on Wikipedia Commons

Colocation vs Cloud Services

The classic make/buy decision revolves around doing things in-house versus outsourcing. For IT managers, this has often come down to a comparison between running an in-house data center and moving everything to a colocation facility. Now there’s a third wrinkle. It’s the cloud.

Both colocation and cloud services have the similarity of being purchased from outside vendors. To the end user, the difference may be trivial. But to someone close to the situation there is a world of difference between colocation and the cloud. Or, is there?

The advantage of colocation facilities is that they provide a physical environment optimized for information technology needs. The colo center is a robust building with rack space all set up with power and cooling. It’s the same support system you would install yourself to support your servers and other data center equipment. You’ll need backup power, of course, fire suppression and physical security. You can’t afford to turn your back on expensive, demanding and possibly temperamental systems, so you need technical staff on duty or nearby. Then there’s the matter of bandwidth connections.

One factor driving many companies to colocation facilities is the need for far greater levels of bandwidth than are available at their current location. This is especially true in smaller towns and rural areas. You need fiber optic services and the best you get is bonded T1. But move to a colo and you’ll be surrounded by carriers who have points of presence in-house and within an easy drop of your equipment cage. You can get all the bandwidth you need, virtually on demand, and at lower prices than you’ll find locally. Why move the whole company when you can simply move a few racks of gear and meet your requirements?

This is the territory that is also now being served by the cloud. The “cloud” is a nebulous (pun intended) term for outsourced information technology services owned and operated by a third party. With a colo center you buy a server and ship it to the colocation operator. In some cases, you rent a server and perhaps the tech support to manage it. But you are responsible for the applications software and operation of the system. When you buy from the cloud, you don’t worry about hardware or software. You are simply purchasing a service. The mechanics behind how that service operates belongs to the cloud service provider.

What can you get from the cloud? Perhaps a specific service. Perhaps everything you need to run your business. This is a new field exhibiting the rapid growth that typifies introduction of a disruptive technology. Buy supercomputing in the cloud when you can’t justify or afford full-time use of a supercomputer on-premises. Get all your office applications through the cloud and say goodbye to software patches and upgrades, not to mention racks of servers in the back room. Perhaps all you need is cloud networking to create a worldwide converged voice and data network for your international business.

What’s best for your business? The answer will very specific to your needs. Perhaps all three approaches will work, but one will have a superior cost/performance characteristic for what you are doing right now. Want some help getting the numbers together? Get quotes for colocation and cloud networking services quickly and easily. Then work through the options and pick what’s optimum for your business.



Note: Photo of clouds and building courtesy of Wikimedia Commons.