Practical Technology

for practical people.

Selling Linux and Open Source to Bean-Counters


I’m no bean-counter. I’m an IT guy. But, I know that over the last few years, it’s CFOs and dollars, not CIOs and gigabytes, that determine what technology companies buy.

But, here’s news you can use to get your CFO on board with a Linux and open-source make over. E-Trade Financial saved $13 million a year and they realized a boost in performance by switching to Linux from Solaris.

Now, $13 million isn’t chicken feed, even to a company like E-Trade that reported $1.7 billion of revenue in 2005.

It wasn’t just Linux that made the difference, though. It was also the Apache Web server and the Jakarta Tomcat JSP (Java Server Pages) servlet system.

In another eWEEK E-Trade story, E-Trade’s VP of architecture, Lee Thompson, explained, “the Red Hat 7.2 kernel came out, which had support for SMP (symmetric multi-processing) and a 32-bit message queue for shared memory. And, all of a sudden, our application booted.”

That, however, was only the first part of the story. It’s what E-Trade did next that many would-be corporate Linux supporters fail at.

Thompson went ahead and. “grabbed a bunch of our architects and we ran like crazy and got a full stack of our application… and we ported a representative stack of our application — our authentication, quote services, product services, some of our trading services and the servlets that rendered the HTML-over to this new stack.”

See the point? Thompson didn’t just show that Linux and open-source could now potentially run the company’s software; they went ahead and showed that it really could run the company’s software.

It’s one thing though to simply run software. It’s another to show that you can do it effectively. Thompson and his crew made that next step.

“We ran some load testing on it, and we knew when it fell over, and the way the Sun systems worked, we could keep adding more and more test users on the Sun box and it would just keep, cranking along — it didn’t really elbow. The Linux box was much faster, and then around, it was somewhere around 180 users, it would elbow… But before 180 users it was much faster.”

So, they showed that Linux was faster, but they also found out where it would stop working effectively. I think it’s very important when trying to sell Linux and open-source in a business to not oversell it.

Time and again, I’ve seen Linux supporters go on and on about how wonderful Linux is, but never mention it’s weaknesses. By doing this, they come off more as fans than serious IT workers. This, in turn, weakens their case.

The final part, and this is where Thompson et. al. showed that they know how modern IT works, is that they then showed that while the Linux systems could only support 180 users, as opposed to the Sun 4500s, which could support 300 to 400 users, the Linux servers cost about $4,000 each while the 4500s were running around a quarter of a million dollars apiece.

Now, those are the kind of numbers, not GHz or Mbps, that make a CFO’s day sunny.

Better still, from the big picture viewpoint, Keynote measurements of transaction numbers showed that with Linux, transactions were only taking 4 to 5 seconds, rather than 8 to 9 seconds with Solaris. For E-trade’s customers — online-traders who want speed, more speed, and then more speed on top of that — this was great.

So what are the lessons from this tale?

First, test out Linux and open-source on the actual jobs your company does. Next, test it with actual corporate work. Then, try it out in production. And, last but never, ever least these days, show how Linux and open-source will make money for your business.

It’s all about the bottom line, and with lower upfront costs than Windows or Solaris, there’s no reason why Linux shouldn’t become THE business operating system of the 21st century.

Leave a Reply