| Which stack is the best stack? |
Jul. 10, 2006
eWEEK Labs has just completed one of the most comprehensive, real-world style tests of Linux, open-source, and .NET application stacks. The results were, from a Linux fan's point of view, annoying.
The tests showed that such vanilla LAMP (Linux, Apache, MySQL, PHP/Python/PERL) stacks as SLES (SUSE Enterprise Linux Server) 9, Zope, ZODB, and PHP and a pure LAMP based on SLES, produced "C" results. They weren't bad, but they weren't anywhere near as good as an out of the box .NET stack based on Windows Server 2003, IIS (Internet Information Server), SQL Server 2005, ASP (Active Server Pages), and SharePoint Portal Server 2003.
Now, having created and run just such tests myself, I know exactly why these benchmarks produced their results. Indeed, eWEEK Labs agrees with me and points these factors out. For example, all their tests were based on standard portal configuration setups. So, you're not really testing the stacks themselves, you're testing the portals.
In addition, none of these setups were optimized. PHP, as any programmer worth their salt knows, is anything but optimized for performance when you first set it up. The .NET stack, however, comes somewhat pre-optimized because it was designed from the start to work together.
Given an expert performance tuner's hand on any of the tested configuration stacks, and you would have seen vastly better results from the Linux-based stacks, and better results from the Windows stacks.
You know what, though? I think the tests were perfectly fair.
Many, perhaps most, IT departments really don't know how to get the most from their software stacks. Oh, they always claim that they know Windows, Linux, or whatever best, and that's why they can't switch to something new. But, when you actually look at what the staff does -- as I have -- you'll find that their expertise is really little more than knowing how to use the most popular, automated scripts.
For all practical purposes, many production systems do end up running on their default settings.
What I and the eWEEK Labs crew found especially interesting, though, was that the combination of open-source components and Server 2003 also produced good results. No one expected that.
The Labs 'WAMP' stacks were made up of Windows Server 2003, Apache, MySQL, and the PHP-based XOOPS; Plone running on Windows Server 2003 R2; and JBoss and MySQL on Windows Server 2003. They did quite well at average transactions per second and download tests. They were, at best however, only OK at average hits per second and in average throughput.
So what should you take away from these sets?
Well, a couple of things. First, Linux and LAMP component developers need to focus on performance. In 1999, when I ran the first objective tests comparing Linux with Microsoft's top Windows server system, then NT, Linux kicked ass and took names.
And today, if you dig into the latest supercomputer rankings, you'll find that most of the world's fastest computers including the top of the top, IBM's Blue Gene/L systems, run Linux. So what's happened?
A big part of it is simply that .NET and Windows are more closely integrated than Linux and all its applications can ever be. The flip-side of the coin, of course, is that if you decide to go with Microsoft, you're stuck with one vendor, and one big and ever-growing price-tag. When it comes to performance for the buck, I'll still go with LAMP any day of the week.
It's not all Microsoft integration though. Those WAMP stacks were pretty darn impressive.
So, I think that a lot of Linux's performance problem is from creeping featuritis. The Linux kernel does a lot more than it used to, and of late, the focus has been more on adding features rather than improving raw performance.
It's not just Linux, though. We could really stand some speed improvements in PHP and the applications that are built on top of it.
I've been continuing to tinker with CMS (Content Management Systems) built on top of PHP, and I've noticed two things.
The first is that there are more than a dozen noteworthy open-source CMSes. That's a lot of wasted duplicative effort. These applications also seem to be all about who has the largest number of features. Performance is not their strong-point.
It's not just CMS. I see many programs being built this way, and I think this trend has gone too far. We need more speed, not more bells and whistles, in our server-side applications.
We also need the Linux distributors to continue to work on their own certified application stacks. For example, Red Hat has started supporting three stacks of its own: LAMP (with the option of PostgreSQL), and a pair of Java-based stacks that are being replaced by a very promising combination of RHEL (Red Had Enterprise Linux) and JBoss. Novell has its own stack based on SLES (SuSE Enterprise Linux Server) and JEMS (JBoss Enterprise Middleware System), which includes the JBoss Application Server, a J2EE (Java 2 Platform, Enterprise Edition) server.
As these distribution-specific stacks mature, I expect to see Linux's overall server application performance increase dramatically. It had better. While for my money, you still can't beat Linux on a bang-for-buck basis, it annoys me to see how inefficient server-side software has gotten in Linux and the open-source world.
If Microsoft can improve its server and applications' speed, why can't the open-source world?
-- Steven J. Vaughan-Nichols
Do you have comments on this story?
Talkback here
NOTE: Please post your comments regarding our articles using the above link. Be sure to use this article's title as the "Subject" in your posts. Before you create a new thread, please check to see if a discussion thread is already running on the article you plan to comment on. Thanks!
(Click here for further information)
|
|
|
7 Advantages of D2D Backup
For decades, tape has been the backup medium of choice. But, now, disk-to-disk (D2D) backup is gaining in favor. Learn why you should make the move in this whitepaper.
4 Legal Reasons to Control Internet Access
The Internet is obviously a valuable resource for many organizations. However, many are exposed to legal liability concerns because they fail to control Internet access. Learn if you're safe in this white paper.
Rapidly Resolve J2EE Application Problems
Whether you are in the process of building J2EE applications or have J2EE applications already running in production, you must ensure that they deliver the expected ROI. Learn how in this white paper.
Load Testing 2.0 for Web 2.0
There are many unknowns in stress testing Web 2.0 applications. Find out how to test the performance of Web 2.0 in this white paper.
Build Better Games Online
For the game infrastructure providers, life is complex. Making money from games has become more complicated. Why? Find out in this white paper.
Building a Virtual Infrastructure from Servers to Storage
This white paper discusses the virtual storage solutions that reduce cost, increase storage utilization, and address the challenges of backing up and restoring Server environments.
Gaining Faster Wireless Connections with WiMAX
Welcome to what is quickly becoming the hyperconnected world where anything that would benefit from being connected to the network will be connected. Learn more in this white paper.
Is Your Desktop a Security Threat?
The new wave of sophisticated crimeware not only targets specific companies, but also targets desktops and laptops as backdoor entryways into those business’ operations and resources. Learn how to stay safe in this white paper.
Increasing SAN Reliability by 100 Percent
Storage area networks (SAN) are a strong part of storage plans. Learn how to increase your reliability and uptime by 100 percent in this case study.
|
|
|
|
|
news feed