Problem solve Get help with specific problems with your technologies, process and projects.

Which switch?

Some guidelines for choosing the right switch manufacturer.

I am often asked which switch manufacturer makes the better switch. There is no one answer to this. While some people purchase based on brand recognition, others purchase on best vendor relationships, while yet another group purchases based on performance issues as they understand them. In an effort to shed some light on the myriad claims, here are the issues that should be considered when evaluating switch equipment.

Cost is the first thing that many consider. This should NEVER be the single factor in determining which switch is proper for your environment. First, there are hidden costs that may not be apparent in the initial quote. You must consider longer term issues such as maintenance, support, warranty periods, performance and functionality, training, expandability and scalability, target size, mean time before failure (MTBF) and, lastly, you need to be able to do some behind the scenes enquiries.

In taking the outline above, we will address each briefly. The first item, maintenance, is more than a maintenance contract. You need to examine the costs associated with the maintenance plan over the life of the switch (see the article on ROI/TCO fruit salad). There are several levels of maintenance available. The plan you chose will be determined by your replacement needs. In short, how long can you be down, or how much screaming are you willing to hear, some would say. Project this over the life of the switch.

Warranty is another misleading comparison. You must figure this for each and every component. For instance, your chassis may have one warranty period, while the components that fit into the chassis as cards may have quite another. Check this for ALL components of your switching solution. If you were to purchase maintenance after the warranty period for the chassis, you may be exposed for a component. This may keep you from having to buy a component out of pocket.

Support – here's a gem! Don't be satisfied if they advertise 24x7. Find out if it works. Talk to other users. Try it for pre-sales support. Find out if it is insourced our outsourced. Ask to make a couple of trial calls. When will on-site be dispatched and what the cost may be. Is on-site insourced or outsourced. These are the folks that will help you when things go wrong. You will do yourself and your company a disservice if this is not correct.

Performance and functionality are probably the two most cloudy issues. One may say they have wire speed performance while the other touts a particular speed. First understand the terminology used. If it is not in a like term where you can compare it to another vendor, ask them to quantify them with a least common denominator to assist you with determining what there speeds are. "Wire speed" sounds good, but certainly needs to be quantified. Also look at buffer and memory. The more the merrier! The next part of this equation is what is offered in the software set. Some will sell you a switch with all software functionality enabled including security. BIG differentiator. If you have to buy each feature set, you run the risk of downtime if you don't get the right one, if it is not operable with what you need it to be (for instance AS/400 SNA, etc.). The incremental costs for the software components can be a budget breaker.

Training is another hidden consideration. If you have to send your staff to classes for weeks or months to learn how to administer the system (plus, an alternate should they want a vacation, there's a concept), Some systems are really more plug and play than others. There are some brilliant minds that work on switches and routers and some not-so brilliant minds. Any good plan will include technology transfer so that your IT staff can handle problems. At the very least, you should have an intelligent pair of hands at the keyboard to work with support. Knowledge is control. You want control!

Expandability and scalability are important if you have a configuration that is expected to do either. This bug has bitten those of us that have been around for years more than once. In order to determine what the true level of a system is, you have to solicit references. This is the only differentiator between a theoretical limit and an actual limit. If they say they can support XX number of ports – as them to prove it. See behind the scenes below.

Target size is a new factor. It is a shame to say, but many companies are beginning to consider this. The agitating elements of our computer world want the biggest impact possible. It is, to them, a coupe to hurt as many end-users as possible. This is not to say that you don't want to buy product from those with market share, but you must consider patch maintenance in your purchase considerations. Learn how many patches are out there. How are these patches applied? How often have they had to be applied? This one is all about covering your bases. If you are okay with doing this weekly or daily, that is fine, just account for the hours in your plan.

Mean time before failure (MTBF) is my favorite peeve. Some people just don't check this figure. When you buy a car, you look at longevity of product, makes sense to do the same here. These figures are generally certified on some basis. I recently helped a customer with a comparison of some switch systems. One of the considerations was MTBF. Their two top vendors had a wide disparity on these calculations. The greater known brand was 1/3 that of the smaller brand. Performance on the less accepted brand was also greater. The proof is in the pudding!

Now for the best trick... behind the scenes investigation. When you get a resume from someone, most of the time you can be assured that the references will say glowing things. It is better to learn from HR. Do a little digging. Don't just take their references. For vendors, check press releases to find out if those who selected the solution are happy. Check companies that make complementary products to find out whose work best, for instance packet shapers, traffic managers, management software, etc. Ask the references you call who else to call. Be sneaky! The extra hours of phone calls may save you hours of headache down the road! Check with the independent testing companies. Find out if the claims are true.

My final recommendation is not to buy what everyone is buying without checking these parameters and finding out if the product will truly serve your needs. In understanding all of the parameters, you will find a much better system for your needs. Pretend this major purchase is a once in a lifetime purchase out of our own pocket. Forget the "Jones's". Remember, your tranquility and time with your family or hobbies count on this one decision, and maybe even your sanity!

Carrie Higbie, Global Network Applications Market Manager, The Siemon Company
Carrie has been involved in the computing and networking industries for nearly 20 years. She has worked with manufacturing firms, medical institutions, casinos, healthcare providers, cable and wireless providers and a wide variety of other industries in both networking design/implementation, project management and software development for privately held consulting firms and most recently Network and Software Solutions.

Carrie currently works with The Siemon Company where her responsibilities include providing liaison services to electronic manufacturers to assure that there is harmony between the active electronics and existing and future cabling infrastructures. She participates with the IEEE, TIA and various consortiums for standards acceptance and works to further educate the end user community on the importance of a quality infrastructure. Carrie currently holds an RCDD/LAN Specialist from BICSI, MCNE from Novell and several other certifications.

This was last published in May 2004

Dig Deeper on Campus area network

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.