Discussions

General J2EE: Competitor to Windows Server 2003

  1. Competitor to Windows Server 2003 (2 messages)

    Now that Microsoft has released Windows Server 2003, it is clear that .NET is the core technology that is used in this product.

    Microsoft is only going to make .NET even bigger in future versions of Windows so the question remains. What is the incentive to use Windows as a platform for hosting J2EE technologies if Windows is fine-tuned and dependent of .NET? From this (http://www.doculabs.com/Downloads/WebServicePerformance_04-03.pdf) Doculabs benchmark study .NET performs 110% better on Windows 2003 than J2EE on Linux.

    In the future, is Java going to be supported soley on Linux/Unix? I think that in order to combat .NET, the major J2EE vendors (Sun, IBM, BEA, Oracle) need to create an operating system (probably based on Linux) which is soley designed for running J2EE applications in which it would perform better than .NET on Windows.

    Another thing to think about is cost. With J2EE (w/ exception of JBoss), most major application servers come in the range of $5-10,000. But with .NET, Windows 2003 Enterprise Edition sells for $3999, and you not only get an easy to use and stable OS, but you also get the high-performance .NET (w/ IIS) application server built-in.

    If the J2EE vendors can somehow create an operating system based on Linux that also combines a high-performance J2EE server, then there is an equal chance of competition against Windows 2003. Otherwise, it seems to me (w/ the exception of JBoss & Linux) that J2EE is more costly and doesn't give as many benefits.

    This is my 2 cents. What do you think?

    Threaded Messages (2)

  2. Addition[ Go to top ]

    By the way, part of IIS is built into the Windows 2003 Kernel:

    CNET: "You pushed some of the IIS into the kernel, didn't you?"
    Windows Kernel Guru Rob Short: "We have what we call a listener, an HTTP handler that we pushed into the kernel. We were looking at how to improve performance. Requests come in and go all the way through the networking and back into user mode where they're handed off. There is a huge amount of the web traffic that you can respond to very quickly without having to have a user mode. So there's HTTP.SYS, a driver that runs in kernel mode and responds in ways that are very well understood, with some parsing and quite a bit of caching, and it handles sessions and it's a huge performance win."

    Source (ZDNET): http://news.zdnet.co.uk/story/0,,t269-s2133899,00.html?rtag=zdnetukhompage

    Maybe part of a J2EE Server could get embedded into the Linux Kernel for it to perform faster.
  3. .NET to Stay[ Go to top ]

    In my opinion you are right on the target regarding the J2EE cost benefit analysis, the dominant platforms are too pricy and if .NET does pickup it will certainly give IT managers across fortune 500 an alternate to pricy Web Sphere and Web Logic application servers.

    However J2EE is not on its way for garbage collection yet. Windows 2003 and .NET integration is not mature yet. The IIS 6.x and .NET integration is good but the framework and infrastructure needed to convenience IT managers is simply not there.

    MS is bound to make that leap with Longhorn "the next MS WIN OS Due 06" by then .NET studios will be matured as well.

    But seriously how can we discount all the stuff happening on organizations such as Eclipse or Fedora etc. Perhaps we all will be pleasantly surprised….

    Thanks,
    Omer Qureshi