Bad News For Ixia, Spirent and Agilent

Rich, first off a disclaimer, I work for a competitor to the testing companies you mention, so I certainly come with an inherent bias I found this an interesting post because with it's brevity it brings up an important topic for the equipment testing industry to pay attention to, particularly in this economic environment.

I'll be interested to hear from other folks using testing equipment, including our own customers, on the value they put on the solutions versus DYI. I've seen a few conversations on this topic in different communities like perftesting.org and even our own blog and imagine it will be one of the first topics for QA and R&D labs throughout 2009.

Looking forward to reading more (and possibly some more pics, was interesting to see their load testing room).

Kyle

Sadly, my tour is over but I can tell you that testing vendors will likely come under pressure in a slower economy. No equipment company or service provider questions the values testing vendors provide but they seem to have sticker shock.

The industry likely needs to change models to a manged services offering with a smaller and more predictable monthly recurring revenue stream.

Well thanks for the insight Rich, I shared your post with a bunch of folks today so it is a topic on the top of our lists.

/kff

Let me declare from the outset that I too work for one of the vendors. The value client derive from the Test Systems is a topic I am very interested in. I would simply like to understand the context, the question being "Expensive, In comparison with what?"

It does say "they use there own equipment when possible" which, I think may mean they use the least cost option to test when it’s technically appropriate, don’t all companies do this? It may also suggest that not all testing can be completed by these systems. In turn that drives a need for the highly complex and comprehensive tools offered by the big two.

I think you’re right that the credit crunch followed by the global recession will have a dramatic effect on the test industry; some won’t make it to the other side, I don’t know right now who will be with us at the dawn of the next upturn.

Test companies have always competed with in-house developed test tools - remember Cisco Pagent (Packet Generator/Analyzer). This does help to drive the cost of test equipment down.
Lease and rental programs also can help companies avoid large CapEx outlays.
In the downturn I think there will be more opportunities for hosted Testing services.
I think the Test companies most at risk are the startups as equipment vendors can't afford to support many different test platforms.
Neal (yes..I work for a leading test company)

As a past consumer of such products, I'll make three points:

1. Yes, it's expensive, but sometimes you just have to pay to play. There's no way around ponying up for the equipment you need to test your own equipment before it ships. Failure to do so is almost always a disaster (speaking from experience).

2. With the rate that old startups shut down and new startups pop up, there seems to be a decent market on ebay and elsewhere for used test equipment.

3. If you're on the bleeding edge, it's hard to find a test equipment vendor who can test what you're building. Or, for the feature set you need, it's a better investment to assign internal resources to build the test equipment to your own custom specs.

I'm not affiliated with any of the test vendors. I do run http://perftesting.org which tries to stay up on performance testing news and such....

I find the general comment that started this thread somewhat lacking in detail. While it is well known that using any of the specialized test gear is expensive, I agree with most of the other posters that you get what you pay for.

If you consider what it costs to develop, in house, not just the tool to perform the testing, but all of the other pieces to maintain and use it, I think you'll find that it's pretty darn expensive to do it yourself. Most folks forget the costs associated with centralized management, centralized statistics, driver and stack optimizations to make things go fast on commodity hardware, and many other aspects of developing things in-house.

I will agree that over the last couple of years, CPUs have become faster and more dense, which allows some of these issues to be overlooked since you can jam more on a single unit. The other main issues still exist even with these improvements.

My company also has our own internal tools, and we use them extensively for internal testing - not final published information.

The most important thing for us is having statistics and results that are completely repeatable by our customers and the rest of the market. You can't do that with your own home-grown app unless you want to publish it, support it, and deal with all of those pieces as well.

I think over the next 2-4 years we will see the CPU increases that have happened in the rest of the market start to affect the performance testing vendors as well. An Atom-based blade or test module is an interesting idea from many angles. Hopefully this combined with the economy as well as the fact that the market is getting more and more saturated and "growing up" a bit will push the per transaction prices down, and open up other models as well.

Test Equipment Rental Buying is very expensive compare to renting. Every project that need test equipment must make their choice either to buy or rent. Project don't last for long so why spend much for equipment if you can have it by renting instead.