Is Linux cheaper than Windows to manage?
As part of its “Get The Facts” campaign, Microsoft has argued for years that Linux is more expensive to manage than Windows. A new study out
this week from Enterprise Management Associates (EMA), sponsored by the
Linux backers OSDL and Levanta, asserts that Linux is now in fact cheaper to
manage than Windows.
Eighty-eight percent of survey respondents reported that they spend less effort
managing Linux than Windows. Some 97 percent noted that, at worst, systems
management effort is the same whether it’s Windows or Linux.
The report also contends that Linux is cheaper to manage than Windows from a
security point of view. 88 percent of respondents indicated that
they spend fewer than 10 minutes per week per server managing malware and
“When asked to compare security management between Linux and Windows, the
interviewees strongly endorsed Linux as easier to manage and inherently
less vulnerable,” the report states. “Only a small minority spend the same
amount of time managing spyware and viruses on Linux and Windows. None
reported spending more time on Linux than Windows.”
Acquisition and resource costs for Linux are apparently lower, as well. EMA
said that Linux acquisition costs may be nearly $60,000 less per server
than Windows. The report also found that Linux is more productive since
Linux system administrators typically handle more systems than Windows
The cost of those Linux systems administrators is also not necessarily that
much higher (if at all) than their Windows counterparts, either. 17 percent
of Windows administrators reported earning more than $70,000 a year, in
comparison to 22 percent for Linux administrators.
administrators earning less than $60,000 per year, 60 percent were Windows
and 52 percent were Linux.
“One of the big issues over time has been that Linux skills were more
expensive to acquire than Windows skills,” OSDL CEO Stuart Cohen told
internetnews.com. “I’m glad that’s gone away.
“Linux has gone mainstream
enough that there are enough people out there that companies can now hire
Linux skills and take advantage of that.”
Though the report identifies Novell SUSE Linux Enterprise Server and Red Hat
Enterprise Linux, (at least in its hardware costs comparison) Cohen doesn’t
think it’s the distribution that matters when it comes to cost of
“I don’t think it’s a given flavor of the distribution that matters, “Cohen
said. “I think it’s more experience level than anything else.”
So what does Microsoft think of the OSDL-sponsored study?
“We’re excited to see the OSDL join Microsoft in working to deliver insights
and facts we know customers need to help inform their IT decisions,”
Microsoft’s Martin Taylor, general manager of Platform Strategy said in
an e-mailed statement.
Taylor noted that there is a need for more collaborative industry research
that delivers a level of transparency in the methodology so customers are
able to apply it in their technology decision-making.
“Microsoft continues to be open to working with partners and competitors,
alike, to jointly commission research that helps all our customers engage in
an informed and respectful debate on the facts,” Taylor said.
The OSDL’s Cohen sees Microsoft’s Get The Facts efforts as
being “just business.”
“I don’t think there was anything offensive about it,” Cohen
said. “Obviously they were
just concerned about Windows keeping the market share that it has. It’s just territorial. It’s just business and I think that’s fine.”
It’s important to note, though, that the EMA study does not specifically
compare total cost of ownership (TCO) of Linux versus Windows.
report focuses on whether server management on Linux is a barrier to
cost-effective operations. Cohen noted that the OSDL has no plans at this
time to commission or sponsor a study on Linux TCO either.
“We didn’t really feel that it was necessary to invest in a lot of research
based on the fact that Linux was already having a lot of success — double-digit, quarter-on-quarter growth for the last 15 quarters,” Cohen said.
Apparently, success speaks for itself.