General discussion

Locked

Evaluating Carrier/Service Provider Offerings

By TTate ·
It is time for my organization to begin evaluating carriers again for data communication services. We have historically had a private global managed service frame-relay network using one of the large carriers. We are expecting to change technologies at this time and want to consider other carriers and their services. What experiences have you encountered with the following carriers/service providers and what technologies (MPLS, private VPN, frame relay, etc.) are you using? What would you recommend to an organization about to go through a network migration?

The carriers/service providers we are considering are:

Savvis
Equant
AT&T
MCI
Virtela

The emphasis at this time needs to be on GLOBAL (US, Canada, Mexico, UK, China, Taiwan, Thailand, Singapore) and on managed services. We do not have the skilled staff necessary to manage these services. So, we depend on managed services to ensure change management, reliability & uptime.

Thanks for your assistance.

This conversation is currently closed to new comments.

3 total posts (Page 1 of 1)  
| Thread display: Collapse - | Expand +

All Comments

Collapse -

Connectivity and Traffic Paths

by addicted2sp33d In reply to Evaluating Carrier/Servic ...

I would recommend that you go with the provider that can keep the most traffic on their hardware and still reach all your locations.

In the event of any network hiccup, having to contact other carriers takes the most-time, and leads to a lot of finger-pointing.

I've never had to run something to Taiwan or China, but I would imagine that getting into China will be very interesting.

From our experience, SAVVIS has been pretty good, but we're only going to Central America. Right now, I wouldn't go with AT&T, because of their pending purchase, simply because anytime something like this happens, all kinds of intangibles get lost.

When XO went into bankruptcy, the company I was with at the time had a few T1's coming in (they were there before I got there)... and for a period of 3 months, we didn't get a single Internet invoice. It was good thing... but it could have just-as-easily turned out badly because it was obvious that our account got lost in the shuffle. Luckily it happened in the accounting department, and not in the provisioning department.

Also, consider how mission-critical some of the connections-are... if they are truly mission critical and cannot withstand ANY downtime without significant operational impacts, consider getting redundant connectivity from different carriers (running on different infrastructure).

Collapse -

Some other concerns to address

by TTate In reply to Connectivity and Traffic ...

I agree about the single carrier if all possible. That is why I believe the players I have listed are capable of those activities. We are currently an AT&T customer, so "getting lost in the shuffle" is the norm jk... they have done pretty well for us and we have had a reliable network for several years. It is just that now it is a time for some changes and want to be sure that we consider all alternatives.

Issues to consider:
1. Change management. We must maintain configurations and documentation showing existing configuration across the enterprise. Changes must be well-managed and documented as part of the network lifecycle and to ensure compliance documentation is maintained. This also includes the ability to remotely manage devices and services.
- Addressing the risk: Use change management tools such as helpdesk for taking and documenting change requests. Other necessary items would include configuration documentation repositories to provide storage for running configurations as well
as OS images for the network devices. Remote (out of band) access will be necessary in some cases to perform maintenance and management. Standard architecture designs are easier to maintain. In addition there should be centralized change management controlling what a site is allowed to expose to the internet using their local connection.

2. Resource management. The network must provide reliable communication services between sites. This means that the performance must be measured and monitored to ensure that the network meets the communication needs of the site.
- Addressing the risk: Implement devices with the ability to track performance and usage information for reporting purposes. Implement network monitoring services to ensure uptime is maintained and downtime is reported as quickly as possible to begin any resolution activities.

3. Contingency management. Network connections are not capable of 100% uptime. There will be issues due to local infrastructure and telco (wiring) problems. Other issues may include weather or other communication interruptions. There should be an alternative means of communication between sites to maintain data services in a secure and managed manner.
- Addressing the risk: Provide a secure alternative communication method for a site to
communicate with the rest of the organization if their primary connection is down for some period of time.

4. Security management. The communication network must be secured from compromise by outside forces as well as secure from unauthorized internal changes. A compromise at one site may cause integrity issues at other sites. Alexander the Great: "Upon the conduct of each depends the fate of all."
- Addressing the risk: Implement a consistent security posture at each network communication point. This is required to ensure the ability to manage, report and reduce risks to the data architecture. A distributed network increases the amount of perimeter services that we have to securely manage and document for compliance purposes.

5. Data integrity. Private data and network services must ensure that the data and
servers will not be exposed to manipulation or capture by unauthorized parties.
- Addressing the risk: Secure the data in transit between facilities and ensure that the endpoints are secure from unauthorized capture and manipulation. Provide secure remote access to internal services for mobile users.

6. Technology skills. There is limited routing knowledge and skills within the organization.
Further training and additional staffing may be required to manage this infrastructure.
- Addressing the risk: Train some additional staff (2-3 people worldwide) in WAN technologies. May need to bring additional staff on board initially to roll out the architecture until it reaches a maturity level requiring little management other than small lifecycle changes.

7. Client communication services. There are services implemented at a Corporate level that will need to be maintained in a distributed environment. Such services include: ftp, secure e-mail (virus/spam-free), internet, data sharing (e.g. EDI and data transmissions), VPN.
- Addressing the risk: This is an unclear issue as to how to manage these services in a such a distributed environment. Some centralized services will be needed, but what, where and how are the questions.

Collapse -

Other OPtions

by nbsbob In reply to Evaluating Carrier/Servic ...

Recently I have started to work with what I am terming "botique"network. One is pure MPLS (I am using them for "near" broadcast qulaity TV Distribution and monitoring and another is an ATM play. The ATM play has been pleasant as their prices are surprisingly competetive and I am so used from years past ATM transport being expensive. Toss me an email and I will send you the companies.

Back to Networks Forum
3 total posts (Page 1 of 1)  

Related Discussions

Related Forums