From owner-freebsd-java Thu Feb 25 16:17:15 1999 Delivered-To: freebsd-java@freebsd.org Received: from ns.mt.sri.com (sri-gw.MT.net [206.127.105.141]) by hub.freebsd.org (Postfix) with ESMTP id 799F314E44 for ; Thu, 25 Feb 1999 16:17:11 -0800 (PST) (envelope-from nate@mt.sri.com) Received: from mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.8/8.8.8) with SMTP id RAA17946; Thu, 25 Feb 1999 17:16:55 -0700 (MST) (envelope-from nate@rocky.mt.sri.com) Received: by mt.sri.com (SMI-8.6/SMI-SVR4) id RAA20100; Thu, 25 Feb 1999 17:16:47 -0700 Date: Thu, 25 Feb 1999 17:16:47 -0700 Message-Id: <199902260016.RAA20100@mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: LODea@vet.com.au (Lachlan O'Dea) Cc: freebsd-java@FreeBSD.ORG Subject: Re: New Volano Report In-Reply-To: <36D5E1CE.44167631@vet.com.au> References: <36D5E1CE.44167631@vet.com.au> X-Mailer: VM 6.34 under 19.16 "Lille" XEmacs Lucid Sender: owner-freebsd-java@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I was just reading the new VolanoMark results (benchmarks of JVM > server-side performance) in this month's JavaWorld. FreeBSD did better > than last time, although it was still the slowest of the bunch. > > The Linux Blackdown JDK was run with TYA, but the author didn't seem to > be aware that TYA will work with FreeBSD too. He didn't realize because it's not widely known. By the time he got around to veryifying his results, it was too late to modify them greatly. I verified the numbers after Alex mentioned it on the list, but I didn't have the correct hardware to run it with either ShuJIT or TYA-JIT. > With a JIT I reckon FreeBSD's results would have been similar to those > of the Blackdown JDK. Also, I thought Blackdown used native threads, > whereas the article says it uses green threads. The Blackdown version can be configured to use green threads or native threads. For almost every application, green threads will outperform the native threads on Linux due to the high overhead of the Linux kernel threads implementation. (Threads in Linux are quite heavy, according to the JDK Linux porting information.) Otherwise it's a good article, and the point about IBM's Java implementation being the fastest is very good for both IBM and Java. Nate To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-java" in the body of the message