Return-Path: X-Original-To: apmail-geronimo-user-archive@www.apache.org Delivered-To: apmail-geronimo-user-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id 5A4B3997F for ; Mon, 25 Jun 2012 14:02:18 +0000 (UTC) Received: (qmail 47638 invoked by uid 500); 25 Jun 2012 14:02:17 -0000 Delivered-To: apmail-geronimo-user-archive@geronimo.apache.org Received: (qmail 47602 invoked by uid 500); 25 Jun 2012 14:02:17 -0000 Mailing-List: contact user-help@geronimo.apache.org; run by ezmlm Precedence: bulk list-help: list-unsubscribe: List-Post: Reply-To: user@geronimo.apache.org List-Id: Delivered-To: mailing list user@geronimo.apache.org Received: (qmail 47590 invoked by uid 99); 25 Jun 2012 14:02:17 -0000 Received: from athena.apache.org (HELO athena.apache.org) (140.211.11.136) by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 25 Jun 2012 14:02:17 +0000 X-ASF-Spam-Status: No, hits=3.2 required=5.0 tests=FSL_FREEMAIL_1,HTML_MESSAGE,RCVD_IN_DNSWL_NONE,SPF_PASS X-Spam-Check-By: apache.org Received-SPF: pass (athena.apache.org: local policy) Received: from [98.138.91.52] (HELO nm14-vm0.bullet.mail.ne1.yahoo.com) (98.138.91.52) by apache.org (qpsmtpd/0.29) with SMTP; Mon, 25 Jun 2012 14:02:09 +0000 Received: from [98.138.90.55] by nm14.bullet.mail.ne1.yahoo.com with NNFMP; 25 Jun 2012 14:01:48 -0000 Received: from [98.138.89.169] by tm8.bullet.mail.ne1.yahoo.com with NNFMP; 25 Jun 2012 14:01:47 -0000 Received: from [127.0.0.1] by omp1025.mail.ne1.yahoo.com with NNFMP; 25 Jun 2012 14:01:47 -0000 X-Yahoo-Newman-Property: ymail-3 X-Yahoo-Newman-Id: 913755.13917.bm@omp1025.mail.ne1.yahoo.com Received: (qmail 4508 invoked by uid 60001); 25 Jun 2012 14:01:42 -0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1340632902; bh=DQ/d1wylOH6jsQuny1x1CKDl5e3ZXP4pVHfvJ/sOwUQ=; h=X-YMail-OSG:Received:X-Mailer:References:Message-ID:Date:From:Reply-To:Subject:To:In-Reply-To:MIME-Version:Content-Type; b=zGVcvms6yfSWEDXomIdepn0R8kP82/T6xlKI8PkbHxnTL8jPqug7wbghpVEpsUUCctpSp/YYZ/w454l8qylK/qyj9YqfCbRnuDDcwHW6CtEcSajsga500jDClcW3gaJrGQfFRRiW2DdR+IjlHntJ1f6Xb3ku2+Ax5gfdytx5+Kc= DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=X-YMail-OSG:Received:X-Mailer:References:Message-ID:Date:From:Reply-To:Subject:To:In-Reply-To:MIME-Version:Content-Type; b=ohO3xGfxVPcXfuE19rJAYTesUfVdtCeBWD/hZtcAKx7jtmFy9FpEbBFmYguR96wLVReEnKP81wkR2eq8E5iIkAADXupxRIyTAVg2Hrvr0/XR/eplah3BRlunV96JzN4P5MTdlmY18SJz/G3VdW80RYsHsOl47qcmQcyG39luRQM=; X-YMail-OSG: S6Uv.sAVM1mKfyAAew4IB7YiGDGBuKuKHBzFOtPL3B8tPsq vhrNXUpGzl8wxTzN6jq3OgLfIC2.2APUJrqb1zJelgpFwTlyzgqkkJR2B5Hy ty5PZ1X1mOWGexiTulLWvnJtuIcFExffSi8TWIUtzuCI2GR8DcsjiXVxY5Fj VhmYCdHN1kaDHaNnN3ezDOnF2iief6Ovqx594s25uXZEB1wpAzpt_DwrJFHe B17TsU0pREMNW79pL48TiQzVcV107Y0exEKiqm1OM9_l9w5y.tqQjQzsZ0nc 7xO_iSH.yT38V2murWHIDL101zQiFHUVH72UPpfDc2_N1QULs2_wXrpWFJEb chF0fqR1RUtEAcJyuWnyGJndlAN70y5d45h_bv3FKyQHpX_QNe5uRMrt44Te CJMQPRjJpFw7x9N7I9NpajvfrH5eOeshk90K26TkxBqDMsivRzS5B0_DMnpw Q4Cr2E4T5Ny7tzpVRZ.C92lWAd6jCrotOgFcZ6W0uMWe_cQ-- Received: from [82.181.39.63] by web125606.mail.ne1.yahoo.com via HTTP; Mon, 25 Jun 2012 07:01:42 PDT X-Mailer: YahooMailWebService/0.8.118.349524 References: <1340539450.9625.YahooMailNeo@web125603.mail.ne1.yahoo.com> <1340632784.98427.YahooMailNeo@web125606.mail.ne1.yahoo.com> Message-ID: <1340632902.3451.YahooMailNeo@web125606.mail.ne1.yahoo.com> Date: Mon, 25 Jun 2012 07:01:42 -0700 (PDT) From: Dok Sander Reply-To: Dok Sander Subject: Fw: Performance between 2.2.1 & 3.0.0 To: "user@geronimo.apache.org" In-Reply-To: <1340632784.98427.YahooMailNeo@web125606.mail.ne1.yahoo.com> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="607277540-442218853-1340632902=:3451" X-Virus-Checked: Checked by ClamAV on apache.org --607277540-442218853-1340632902=:3451 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable =0A=0A=0A----- Forwarded Message -----=0AFrom: Dok Sander =0ATo: Forrest Xia ; "user@geronimo.apache.org" =0ASent: Monday, June 25, 2012 4:59 PM=0ASubject: = Re: Performance between 2.2.1 & 3.0.0=0A =0A=0AHi!=0A=0AYes, I don't really= see the startup time as an issue(Since it wasn't =0Adrastically longer bet= ween the versions, IMO) but the operation time of =0Athe deploy command/mod= ule and overall portlet latency should be shorter. I could add my findings = to the issue 6222.=0A=0A=0A=0A--vdox=0A=0A=0A______________________________= __=0A From: Forrest Xia =0ATo: user@geronimo.apache.or= g; Dok Sander =0ASent: Monday, June 25, 2012 7:01 AM= =0ASubject: Re: Performance between 2.2.1 & 3.0.0=0A =0A=0AHi vdox,=0A=0ATh= ank you for your testing on server startup and deployment time performance.= I think they are useful information for the community to trigger some anal= ysis. But before that, I also did some testing on my linux box(Ubuntu 10.04= x86_64). Here are my configuration and results:=0A=0A$ uname -a=0ALinux fo= rrestxm-u64 2.6.32-41-generic #90-Ubuntu SMP Tue May 22 11:29:51 UTC 2012 x= 86_64 GNU/Linux=0A=0A$ free -t=0A=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 total= =A0=A0=A0=A0=A0=A0 used=A0=A0=A0=A0=A0=A0 free=A0=A0=A0=A0 shared=A0=A0=A0 = buffers=A0=A0=A0=A0 cached=0AMem:=A0=A0=A0=A0=A0=A0 3608260=A0=A0=A0 285688= 4=A0=A0=A0=A0 751376=A0=A0=A0=A0=A0=A0=A0=A0=A0 0=A0=A0=A0 1196196=A0=A0=A0= =A0 735504=0A-/+ buffers/cache:=A0=A0=A0=A0 925184=A0=A0=A0 2683076=0ASwap:= =A0=A0=A0=A0=A0 4305912=A0=A0=A0=A0=A0 40024=A0=A0=A0 4265888=0ATotal:=A0= =A0=A0=A0 7914172=A0=A0=A0 2896908=A0=A0=A0 5017264=0A=0A$ cat /proc/cpuinf= o =0Aprocessor=A0=A0=A0 : 0=0Avendor_id=A0=A0=A0 : GenuineIntel=0Acpu famil= y=A0=A0=A0 : 6=0Amodel=A0=A0=A0 =A0=A0=A0 : 42=0Amodel name=A0=A0=A0 : Inte= l(R) Core(TM) i5-2540M CPU @ 2.60GHz=0Astepping=A0=A0=A0 : 8=0Acpu MHz=A0= =A0=A0 =A0=A0=A0 : 2601.000=0Acache size=A0=A0=A0 : 3072 KB=0Aphysical id= =A0=A0=A0 : 0=0Asiblings=A0=A0=A0 : 2=0Acore id=A0=A0=A0 =A0=A0=A0 : 1=0Acp= u cores=A0=A0=A0 : 1=0Aapicid=A0=A0=A0 =A0=A0=A0 : 2=0Ainitial apicid=A0=A0= =A0 : 2=0Afpu=A0=A0=A0 =A0=A0=A0 : yes=0Afpu_exception=A0=A0=A0 : yes=0Acpu= id level=A0=A0=A0 : 13=0Awp=A0=A0=A0 =A0=A0=A0 : yes=0Aflags=A0=A0=A0 =A0= =A0=A0 : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat = pse36 clflush dts acpi mmx fxsr sse sse2 ss syscall nx lm constant_tsc arch= _perfmon pebs bts rep_good xtopology tsc_reliable nonstop_tsc aperfmperf pn= i ds_cpl ssse3 cx16 popcnt lahf_lm ida arat=0Abogomips=A0=A0=A0 : 5202.00= =0Aclflush size=A0=A0=A0 : 64=0Acache_alignment=A0=A0=A0 : 64=0Aaddress siz= es=A0=A0=A0 : 36 bits physical, 48 bits virtual=0Apower management:=0A=0Apr= ocessor=A0=A0=A0 : 1=0Avendor_id=A0=A0=A0 : GenuineIntel=0Acpu family=A0=A0= =A0 : 6=0Amodel=A0=A0=A0 =A0=A0=A0 : 42=0Amodel name=A0=A0=A0 : Intel(R) Co= re(TM) i5-2540M CPU @ 2.60GHz=0Astepping=A0=A0=A0 : 8=0Acpu MHz=A0=A0=A0 = =A0=A0=A0 : 2601.000=0Acache size=A0=A0=A0 : 3072 KB=0Aphysical id=A0=A0=A0= : 0=0Asiblings=A0=A0=A0 : 2=0Acore id=A0=A0=A0 =A0=A0=A0 : 1=0Acpu cores= =A0=A0=A0 : 1=0Aapicid=A0=A0=A0 =A0=A0=A0 : 3=0Ainitial apicid=A0=A0=A0 : 3= =0Afpu=A0=A0=A0 =A0=A0=A0 : yes=0Afpu_exception=A0=A0=A0 : yes=0Acpuid leve= l=A0=A0=A0 : 13=0Awp=A0=A0=A0 =A0=A0=A0 : yes=0Aflags=A0=A0=A0 =A0=A0=A0 : = fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clf= lush dts acpi mmx fxsr sse sse2 ss syscall nx lm constant_tsc arch_perfmon = pebs bts rep_good xtopology tsc_reliable nonstop_tsc aperfmperf pni ds_cpl = ssse3 cx16 popcnt lahf_lm ida arat=0Abogomips=A0=A0=A0 : 5202.00=0Aclflush = size=A0=A0=A0 : 64=0Acache_alignment=A0=A0=A0 : 64=0Aaddress sizes=A0=A0=A0= : 36 bits physical, 48 bits virtual=0Apower management:=0A=0A$ java -versi= on=0Ajava version "1.6.0_29"=0AJava(TM) SE Runtime Environment (build 1.6.0= _29-b11)=0AJava HotSpot(TM) 64-Bit Server VM (build 20.4-b02, mixed mode)= =0A=0A$ export GERONIMO_OPTS=3D"-Djava.security.egd=3Dfile:/dev/./urandom"= =0A=0AG 2.2.1=0A1. First time startup: Startup completed in 16.641s seconds= =0A2. Second time startup: Startup completed in 14.357s seconds=0A=0AG 3.0.= 0-SNAPSHOT built on 20120620:=0A1. First time startup: Startup completed in= 21.995s seconds=0A2. Second time startup: Startup completed in 15.935s sec= onds=0A=0ASo from the startup time results of mine, there is no big gap bet= ween 2.2.1 and 3.0.0-SNAPSHOT. But, for deployment time, there are some gap= s saw from my results as well. =0A=0AAbout the deployment time issue, there= is a known jira opened (https://issues.apache.org/jira/browse/GERONIMO-622= 2), also there are some improvements implemented in https://issues.apache.o= rg/jira/browse/GERONIMO-5813.=0A=0A=0AOn Sun, Jun 24, 2012 at 8:04 PM, Dok = Sander wrote:=0A=0AHello,=0A>=0A>=A0This was discusse= d with kevan to some extend on geronimo IRC channel on Freenode.=0A>=0A>The= re seems to be a considerable performance hit between Geronimo 2.2.1 and 3.= 0.0. In addition to results below the overall loading times of the web UI a= re much longer in 3.0.0 than in 2.2.1. I understand this is a good time to = try and find if the performance can be made better since the actual release= of 3.0.0 is drawing near.=0A>=0A>=0A>=0A>-- vdox=0A>=0A>=0A>=0A>Platform s= pecs:=0A>=0A># uname -s -r -v -m -p -i -o=0A>Linux 3.1.10-1.9-ec2 #1 SMP Th= u Apr 5 18:48:38 UTC 2012 (4a97ec8) x86_64 x86_64 x86_64 GNU/Linux=0A>=0A>#= free -t=0A>=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 total=A0=A0=A0=A0=A0=A0 us= ed=A0=A0=A0=A0=A0=A0 free=A0=A0=A0=A0 shared=A0=A0=A0 buffers=A0=A0=A0=A0 c= ached=0A>Mem:=A0=A0=A0=A0=A0=A0 1751644=A0=A0=A0 1385060=A0=A0=A0=A0 366584= =A0=A0=A0=A0=A0=A0=A0=A0=A0 0=A0=A0=A0=A0=A0 64644=A0=A0=A0=A0 827940=0A>-/= + buffers/cache:=A0=A0=A0=A0 492476=A0=A0=A0 1259168=0A>Swap:=A0=A0=A0=A0= =A0=A0=A0=A0=A0=A0=A0 0=A0=A0=A0=A0=A0=A0=A0=A0=A0 0=A0=A0=A0=A0=A0=A0=A0= =A0=A0 0=0A>Total:=A0=A0=A0=A0 1751644=A0=A0=A0 1385060=A0=A0=A0=A0 366584= =0A>=0A># cat /proc/cpuinfo=0A>processor=A0=A0=A0=A0=A0=A0 : 0=0A>vendor_id= =A0=A0=A0=A0=A0=A0 : GenuineIntel=0A>cpu family=A0=A0=A0=A0=A0 : 6=0A>model= =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 : 26=0A>model name=A0=A0=A0=A0=A0 : Intel(R)= Xeon(R) CPU=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 E5507=A0 @ 2.27GHz=0A>stepping= =A0=A0=A0=A0=A0=A0=A0 : 5=0A>cpu MHz=A0=A0=A0=A0=A0=A0=A0=A0 : 2266.746=0A>= cache size=A0=A0=A0=A0=A0 : 4096 KB=0A>fpu=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0= =A0 : yes=0A>fpu_exception=A0=A0 : yes=0A>cpuid level=A0=A0=A0=A0 : 11=0A>w= p=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 : yes=0A>flags=A0=A0=A0=A0=A0=A0= =A0=A0=A0=A0 : fpu de tsc msr pae cx8 apic sep cmov pat clflush acpi mmx fx= sr sse sse2 ss ht syscall nx lm constant_tsc up rep_good nopl nonstop_tsc a= perfmperf pni ssse3 cx16 sse4_1 sse4_2 popcnt hypervisor lahf_lm dts=0A>bog= omips=A0=A0=A0=A0=A0=A0=A0 : 4555.21=0A>clflush size=A0=A0=A0 : 64=0A>cache= _alignment : 64=0A>address sizes=A0=A0 : 40 bits physical, 48 bits virtual= =0A>power management:=0A>=0A># java -version=0A>java version "1.7.0_b147-ic= edtea"=0A>OpenJDK Runtime Environment (suse-1.2-x86_64)=0A>OpenJDK 64-Bit S= erver VM (build 22.0-b10, mixed mode)=0A>=0A>=0A>-------------=0A>=0A>Test = results:=0A>=0A>=0A>Following JAVA_OPTS are used for both versions:=0A>-Xmx= 512m -Xms256m -XX:MaxPermSize=3D256m=0A>=0A>=0A>=0A>geronimo-tomcat6-javaee= 5-2.2.1=0A>________________________________________________________________= ___________________________________________________________=0A>=0A>=0A>star= tup(From geronimo.out):=0A>Startup completed in 1:46.803s seconds=0A>=0A>li= st modules=0A># time /$GERONIMO_HOME/bin/deploy.sh -u system -p manager lis= t-modules=0A>...=0A>real=A0=A0=A0 0m13.703s=0A>user=A0=A0=A0 0m9.617s=0A>sy= s=A0=A0=A0=A0 0m0.648s=0A>=0A>=0A>deploy=0A># time $GERONIMO_HOME/bin/deplo= y.sh -u system -p manager deploy --inPlace=0A>...=0A>real=A0=A0=A0 0m12.815= s=0A>user=A0=A0=A0 0m9.137s=0A>sys=A0=A0=A0=A0 0m0.740s=0A>=0A>undeploy:=0A= ># time $GERONIMO_HOME/bin/deploy.sh -u system -p manager undeploy org.apac= he.geronimo.samples/HelloWorldApp/1.0/war=0A>...=0A>real=A0=A0=A0 0m13.562s= =0A>user=A0=A0=A0 0m10.457s=0A>sys=A0=A0=A0=A0 0m0.840s=0A>=0A>=0A>CPU usag= e during deploy list-modules:=0A>geronimo: 0-20%=0A>deploy: 50-90%=0A>_____= ___________________________________________________________________________= ___________________________________________=0A>=0A>=0A>=0A>=0A>=0A>geronimo= -tomcat7-javaee6-3.0.03.0.0-20120620.212946 (I originally tested with 3.0.0= -beta1 but snapshot was recommended to me by kevan)=0A>____________________= ___________________________________________________________________________= ____________________________=0A>=0A>startup(From geronimo.out):=0A>"Startup= completed in 2:20.080s seconds"=0A>=0A>=0A>list modules:=0A>time $GERONIMO= _HOME/bin/deploy -u system -p manager list-modules=0A>...=0A>real=A0=A0=A0 = 0m51.848s=0A>user=A0=A0=A0 0m36.230s=0A>sys=A0=A0=A0=A0 0m3.376s=0A>=0A>=0A= >deploy:=0A>time $GERONIMO_HOME/bin/deploy -u system -p manager deploy --in= Place .=0A>...=0A>real=A0=A0=A0 0m56.284s=0A>user=A0=A0=A0 0m32.822s=0A>sys= =A0=A0=A0=A0 0m3.680s=0A>=0A>=0A>undeploy:=0A>time $GERONIMO_HOME/bin/deplo= y -u system -p manager undeploy org.apache.geronimo.samples/HelloWorldApp/1= .0/war=0A>...=0A>real=A0=A0=A0 0m47.336s=0A>user=A0=A0=A0 0m32.862s=0A>sys= =A0=A0=A0=A0 0m2.764s=0A>=0A>=0A>=0A>CPU usage during deploy list-modules:= =0A>geronimo: 80-90%=0A>deploy: 10-20%=0A>=0A>_____________________________= ___________________________________________________________________________= ___________________=0A>=0A=0A=0A-- =0AThanks!=0A=0ARegards, Forrest --607277540-442218853-1340632902=:3451 Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable

----- Forwarded Message -----
From: Dok Sander <doksander@yahoo.com> To: Forrest Xia <forr= estxm@gmail.com>; "user@geronimo.apache.org" <user@geronimo.apache.or= g>
Sent: Monday, J= une 25, 2012 4:59 PM
Subject: Re: Performance between 2.2.1 & 3.0.0

= =0A
<= div>Hi!

Yes, I=0A = don't really see the startup time as an issue(Since it wasn't =0Adrasticall= y longer between the versions, IMO) but the operation time of =0Athe deploy= command/module and overall portlet latency should be shorter.=0A I could a= dd my findings to the issue 6222.


--vdox

From: Forrest= Xia <forrestxm@gmail.com>
T= o: user@geronimo.apache.org; Dok Sander <doksander@yahoo.com&= gt;
Sent: Monday, June= 25, 2012 7:01 AM
Subject:<= /b> Re: Performance between 2.2.1 & 3.0.0

=0AHi vdox,

Thank you for your testing on server = startup and deployment time performance. I think they are useful informatio= n for the community to trigger some analysis. But before that, I also did s= ome testing on my linux box(Ubuntu 10.04 x86_64). Here are my configuration= and results:
=0A
$ uname -a
Linux forrestxm-u64 2.6.32-41-generic= #90-Ubuntu SMP Tue May 22 11:29:51 UTC 2012 x86_64 GNU/Linux

$ free= -t
           &n= bsp; total       used    =    free     shared    buffers&= nbsp;    cached
Mem:       = 3608260    2856884     751376  = ;        0    1196196&nbs= p;    735504
=0A-/+ buffers/cache:    = ; 925184    2683076
Swap:      4= 305912      40024    4265888
Tot= al:     7914172    2896908  &n= bsp; 5017264

$ cat /proc/cpuinfo
processor    : 0=
vendor_id    : GenuineIntel
cpu family  &nb= sp; : 6
=0Amodel        : 42
model name=     : Intel(R) Core(TM) i5-2540M CPU @ 2.60GHz
stepping&n= bsp;   : 8
cpu MHz        : 2601= .000
cache size    : 3072 KB
physical id  &n= bsp; : 0
siblings    : 2
core id    &nb= sp;   : 1
cpu cores    : 1
=0Aapicid &n= bsp;      : 2
initial apicid    : 2fpu        : yes
fpu_exception &nbs= p;  : yes
cpuid level    : 13
wp   = ;     : yes
flags        : = fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clf= lush dts acpi mmx fxsr sse sse2 ss syscall nx lm constant_tsc arch_perfmon = pebs bts rep_good xtopology tsc_reliable nonstop_tsc aperfmperf pni ds_cpl = ssse3 cx16 popcnt lahf_lm ida arat
=0Abogomips    : 5202.= 00
clflush size    : 64
cache_alignment  &nb= sp; : 64
address sizes    : 36 bits physical, 48 bits vir= tual
power management:

processor    : 1
vendor_= id    : GenuineIntel
cpu family    : 6
= =0Amodel        : 42
model name  = ;  : Intel(R) Core(TM) i5-2540M CPU @ 2.60GHz
stepping  &= nbsp; : 8
cpu MHz        : 2601.000
cac= he size    : 3072 KB
physical id    : 0siblings    : 2
core id      &n= bsp; : 1
cpu cores    : 1
=0Aapicid    =     : 3
initial apicid    : 3
fpu =        : yes
fpu_exception    : = yes
cpuid level    : 13
wp     &nb= sp;  : yes
flags        : fpu vme de = pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts ac= pi mmx fxsr sse sse2 ss syscall nx lm constant_tsc arch_perfmon pebs bts re= p_good xtopology tsc_reliable nonstop_tsc aperfmperf pni ds_cpl ssse3 cx16 = popcnt lahf_lm ida arat
=0Abogomips    : 5202.00
clflu= sh size    : 64
cache_alignment    : 64address sizes    : 36 bits physical, 48 bits virtual
pow= er management:

$ java -version
java version "1.6.0_29"
Java(TM= ) SE Runtime Environment (build 1.6.0_29-b11)
=0AJava HotSpot(TM) 64-Bit= Server VM (build 20.4-b02, mixed mode)

$ export GERONIMO_OPTS=3D"-D= java.security.egd=3Dfile:/dev/./urandom"

G 2.2.1
1. First time st= artup: Startup completed in 16.641s seconds
=0A2. Second time startup: S= tartup completed in 14.357s seconds

G 3.0.0-SNAPSHOT built on 201206= 20:
1. First time startup: Startup completed in 21.995s seconds
2. Se= cond time startup: Startup completed in 15.935s seconds
=0A
So from t= he startup time results of mine, there is no big gap between 2.2.1 and 3.0.= 0-SNAPSHOT. But, for deployment time, there are some gaps saw from my resul= ts as well.

About the deployment time issue, there is a known jira = opened (https://issues.apache.org/jira/browse/GERON= IMO-6222), also there are some improvements implemented in https://issues.apache.org/jira/browse/GERONIMO-5813.
=0A=
On Sun, Jun 24, 2012 at 8:04 PM,= Dok Sander <dok= sander@yahoo.com> wrote:
=0AHello,
=0A
=0A This was discussed with kevan to so= me extend on geronimo IRC channel on Freenode.
=0A
=0AThere seems to = be a considerable performance hit between Geronimo 2.2.1 and 3.0.0. In addi= tion to results below the overall loading times of the web UI are much long= er in 3.0.0 than in 2.2.1. I understand this is a good time to try and find= if the performance can be made better since the actual release of 3.0.0 is= drawing near.
=0A=0A
=0A
=0A
=0A-- vdox
=0A
=0A
=0A=0APlatform specs:
=0A
=0A# uname -s -r -v -m -p -i -o
=0ALinux = 3.1.10-1.9-ec2 #1 SMP Thu Apr 5 18:48:38 UTC 2012 (4a97ec8) x86_64 x86_64 x= 86_64 GNU/Linux
=0A
=0A# free -t
=0A     =         total    &nb= sp;  used       free   &n= bsp; shared    buffers     cached
=0A= Mem:       1751644    1385060&= nbsp;    366584       &nb= sp;  0      64644     827= 940
=0A-/+ buffers/cache:     492476  &nbs= p; 1259168
=0ASwap:         = ;   0          0&nbs= p;         0
=0ATotal: &nbs= p;   1751644    1385060     36= 6584
=0A
=0A# cat /proc/cpuinfo
=0Aprocessor   &nbs= p;   : 0
=0Avendor_id       : Ge= nuineIntel
=0Acpu family      : 6
=0Amodel&n= bsp;          : 26
=0Amodel= name      : Intel(R) Xeon(R) CPU   = ;        E5507  @ 2.27GHz
=0Aste= pping        : 5
=0Acpu MHz &nbs= p;       : 2266.746
=0Acache size &nb= sp;    : 4096 KB
=0Afpu     &nbs= p;       : yes
=0Afpu_exception  = ; : yes
=0Acpuid level     : 11
=0Awp  =             : yes=0Aflags           : fpu= de tsc msr pae cx8 apic sep cmov pat clflush acpi mmx fxsr sse sse2 ss ht = syscall nx lm constant_tsc up rep_good nopl nonstop_tsc aperfmperf pni ssse= 3 cx16 sse4_1 sse4_2 popcnt hypervisor lahf_lm dts
=0A=0Abogomips &= nbsp;      : 4555.21
=0Aclflush size  = ;  : 64
=0Acache_alignment : 64
=0Aaddress sizes   : 4= 0 bits physical, 48 bits virtual
=0Apower management:
=0A
=0A# jav= a -version
=0Ajava version "1.7.0_b147-icedtea"
=0AOpenJDK Runtime En= vironment (suse-1.2-x86_64)
=0AOpenJDK 64-Bit Server VM (build 22.0-b10,= mixed mode)
=0A
=0A
=0A-------------
=0A
=0ATest results:=0A
=0A
=0AFollowing JAVA_OPTS are used for both versions:
=0A-X= mx512m -Xms256m -XX:MaxPermSize=3D256m
=0A
=0A
=0A
=0Ageronimo-= tomcat6-javaee5-2.2.1
=0A_______________________________________________= ___________________________________________________________________________= _
=0A
=0A
=0Astartup(From geronimo.out):
=0AStartup completed i= n 1:46.803s seconds
=0A
=0Alist modules
=0A# time /$GERONIMO_HOME/= bin/deploy.sh -u system -p manager list-modules
=0A...
=0Areal &= nbsp;  0m13.703s
=0Auser    0m9.617s
=0Asys =     0m0.648s
=0A
=0A
=0Adeploy
=0A# time $GERONI= MO_HOME/bin/deploy.sh -u system -p manager deploy --inPlace
=0A...
= =0Areal    0m12.815s
=0Auser    0m9.137s=0Asys     0m0.740s
=0A
=0Aundeploy:
=0A# ti= me $GERONIMO_HOME/bin/deploy.sh -u system -p manager undeploy org.apache.ge= ronimo.samples/HelloWorldApp/1.0/war
=0A...
=0Areal   = 0m13.562s
=0Auser    0m10.457s
=0Asys  &nbs= p;  0m0.840s
=0A
=0A
=0ACPU usage during deploy list-modules:=
=0Ageronimo: 0-20%
=0Adeploy: 50-90%
=0A_________________________= ___________________________________________________________________________= _______________________
=0A
=0A
=0A
=0A
=0A
=0Ageronimo-t= omcat7-javaee6-3.0.03.0.0-20120620.212946 (I originally tested with 3.0.0-b= eta1 but snapshot was recommended to me by kevan)
=0A___________________= ___________________________________________________________________________= _____________________________
=0A
=0Astartup(From geronimo.out):
= =0A"Startup completed in 2:20.080s seconds"
=0A
=0A
=0Alist module= s:
=0Atime $GERONIMO_HOME/bin/deploy -u system -p manager list-modules=0A...
=0Areal    0m51.848s
=0Auser   = ; 0m36.230s
=0Asys     0m3.376s
=0A
=0A
=0A= deploy:
=0Atime $GERONIMO_HOME/bin/deploy -u system -p manager deploy --= inPlace .
=0A...
=0Areal    0m56.284s
=0Auser =    0m32.822s
=0Asys     0m3.680s
=0A=0A
=0Aundeploy:
=0Atime $GERONIMO_HOME/bin/deploy -u system -p mana= ger undeploy org.apache.geronimo.samples/HelloWorldApp/1.0/war
=0A...=0Areal    0m47.336s
=0Auser    0m32.862s=
=0Asys     0m2.764s
=0A
=0A
=0A
=0ACPU = usage during deploy list-modules:
=0Ageronimo: 80-90%
=0Adeploy: 10-2= 0%
=0A
=0A___________________________________________________________= ________________________________________________________________
=0A



--
Thanks!

Regards, For= rest

=0A




--607277540-442218853-1340632902=:3451--