From owner-freebsd-performance@FreeBSD.ORG Sun May 30 06:39:20 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id A769516A4CE; Sun, 30 May 2004 06:39:20 -0700 (PDT) Received: from p4.roq.com (ns1.ecoms.com [207.44.130.137]) by mx1.FreeBSD.org (Postfix) with ESMTP id 1961143D31; Sun, 30 May 2004 06:39:20 -0700 (PDT) (envelope-from michael@roq.com) Received: from roq.com (CPE-203-51-131-166.vic.bigpond.net.au [203.51.131.166]) by p4.roq.com (Postfix) with ESMTP id 017174C563; Sun, 30 May 2004 13:39:20 +0000 (GMT) Message-ID: <40B9E3F8.1020301@roq.com> Date: Sun, 30 May 2004 23:39:04 +1000 From: Michael Vince User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:1.6) Gecko/20040520 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Vasenin Alexander aka BlackSir References: In-Reply-To: X-Enigmail-Version: 0.83.6.0 X-Enigmail-Supports: pgp-inline, pgp-mime Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit cc: freebsd-isp@freebsd.org cc: freebsd-performance@freebsd.org cc: Maksymilian Wrzesinski Subject: Re: raid backup X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 30 May 2004 13:39:20 -0000 Vasenin Alexander aka BlackSir wrote: > I'm watched nearly same thing yesterday and found, that all cheap solutions is the so-called HostRAID, that are software or > semi-software implementations of RAID. So, their "RAID" ability supported only by M$ Window$. Under FreeBSD they work as normal > IDE/SCSI/SATA controllers. I've tested yesterday one of them - Adaptec on AIC 7899 chip. Take a look at hardware notes on 4.10, for > example "Promise Fasttrak-33, -66, -100, -100 TX2/TX4, -133 TX2/TX2000" - noone except Promise calls it RAID... > Vasenin Alexander aka BlackSir Yep thats what I have largely thought until I started to come across things like this. http://www.techno-obscura.com/~delgado/notes/fbsd-ich5SATAraid.html Using the atacontrol he cliams its "sort of hot swappable" I dont know if you can call it hardware raid but it certainly sounds hardware raidish enought for me to be happy with it... "atacontrol is a tool that has been kicking around in FreeBSD for a while, but I hadn't tried it because it seemed to be directed at managing other hardware ATA RAID setups. However, it ends up that this works like a charm (just about) with the built in goods." >>-----Original Message----- >>From: owner-freebsd-isp@freebsd.org [mailto:owner-freebsd-isp@freebsd.org]On Behalf Of Michael Vince >>Sent: Saturday, May 29, 2004 2:03 PM >>Cc: freebsd-isp@freebsd.org; freebsd-performance@freebsd.org; Maksymilian Wrzesinski >>Subject: Re: raid backup >> >>Looking around on the net and FreeBSD archives it looks like if you want >>cheap ATA Raid now go with either a Intel motherboard with ICH5 or >>Promise SATA TX2/TX4 >>The performance of the Intel ICH5 looks very good with low CPU >>utilization (well from what benchmarks in running from MS windows suggest) >>I haven't found any clear documentation on what can and cant be used. > > From owner-freebsd-performance@FreeBSD.ORG Tue Jun 1 16:03:09 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id EC50716A4CE for ; Tue, 1 Jun 2004 16:03:08 -0700 (PDT) Received: from beelzebubba.sysabend.org (alcatraz.inna.net [209.201.74.14]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6207F43D31 for ; Tue, 1 Jun 2004 16:03:08 -0700 (PDT) (envelope-from xyzzy@moo.sysabend.org) Received: from moo.sysabend.org (moo.sysabend.org [66.111.41.70]) by beelzebubba.sysabend.org (Postfix) with SMTP id 50F6914F15 for ; Tue, 1 Jun 2004 19:03:20 -0400 (EDT) Received: (nullmailer pid 21738 invoked by uid 14); Tue, 01 Jun 2004 23:03:06 -0000 Date: Tue, 1 Jun 2004 16:03:06 -0700 From: Tom Arnold To: performance@FreeBSD.ORG Message-ID: <20040601230306.GN9728@moo.sysabend.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.4i Organization: The Sysabend Dump X-Operating-System: CPM2.2 X-Bucket-Brigade-Devices: Rah! X-8-Bit-Samples-And-Analog-Filters: Rah! Subject: Mysql Performance issues on BSD4.9 X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: xyzzy@sysabend.org List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 01 Jun 2004 23:03:09 -0000 I'll try to give background and be as brief as possible. Hardware : Dell 2650, dual 2.4ghz Xeon, 3gigs RAM. Original configuration BSD4.8 Raid 1+0 ( or is it 0+1 ) using onboard Perc3 controller. ( 4 70gig drives ) plus an additional 70gig drive in the 5th slot. Raid was full so we added the 5th drive and expanded the Mysql database onto it. Speed was "tolerable" New configuration BSD4.9 5x140gig drives ( 15k LVD ) on same Perc3 controller. Raid5 - Way incredably painfully slow. Slave DB would take days to catch up with Master. Raid0+1( or is it 1+0 damn adaptec ) + 3 seperate disks. All on Perc3 controller. - Still slow. We've used the pair of mirrored disks as the boot disks and the mysql DB is broken up across the 3 seperates. The machine is still sluggish. I'll paste a systat -v at the end of the message. I'm going to apologise right off with not being more detailed about what I mean by "slow" and "sluggish". If you can tell me more detail to look for I'll be happy to supply. Now, my theories for you to poke holes in. 1. The original configuration was speed-tolerable because it was basically one large logical drive ( ignoring the standalone drive we added ) and it was smaller drives. 2. The RAID5 configuration was dirt-slow because its Raid5 and you take a large performance hit ( 4X normal write time? ). 3. The RAID10+3XStandalone is slow because now we have 4 seperate drives ( 1 logical 3 physical ) going via the same controller. 4. Optimal speed for this would be, say, 14 smaller ( 36 or 70gig ) drives in RAID10. This is kind of my silver bullet solution based on experiences running INND all those years ago. Here's a Systat -v from the Raid10 + 3 seperate drive configuration. Thanks! Mem:KB REAL VIRTUAL VN PAGER SWAP PAGER Tot Share Tot Share Free in out in out Act 1222516 2156 1303312 3476 109292 count All 3081632 4680 3870300 8488 pages 2085 zfod Interrupts Proc:r p d s w Csw Trp Sys Int Sof Flt cow 3824 total 20 2105 12916 382639428 3824 5984 2109 252952 wire ata0 irq14 1239708 act 2881 bge0 irq11 15.8%Sys 1.5%Intr 1.2%User 19.6%Nice 61.9%Idl 1484452 inact 486 aac0 irq16 | | | | | | | | | | 104520 cache fdc0 irq6 ========+---------- 4772 free atkbd0 irq daefr 201 clk irq0 Namei Name-cache Dir-cache prcfr 256 rtc irq8 Calls hits % hits % react 72 72 100 pdwake pdpgs Disks aacd0 aacd1 aacd2 aacd3 acd0 fd0 md0 intrn KB/t 59.64 64.00 64.00 64.00 0.00 0.00 0.00 197120 buf tps 11 97 76 61 0 0 0 145 dirtybuf MB/s 0.63 6.05 4.76 3.79 0.00 0.00 0.00 199064 desiredvnodes % busy 1 56 57 39 0 0 0 49766 numvnodes 37974 freevnodes -- ------------------------------------------------------------------------ - Tom Arnold - When I was small, I was in love, - - Sysabend - In love with everything. - - CareTaker - And now there's only you... - -------------- -- Thomas Dolby, "Cloudburst At Shingle Street" - From owner-freebsd-performance@FreeBSD.ORG Wed Jun 2 23:44:10 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E299D16A4CE for ; Wed, 2 Jun 2004 23:44:10 -0700 (PDT) Received: from smtp14.singnet.com.sg (smtp14.singnet.com.sg [165.21.6.24]) by mx1.FreeBSD.org (Postfix) with ESMTP id ABD2443D48 for ; Wed, 2 Jun 2004 23:44:09 -0700 (PDT) (envelope-from shanali@blanc.magix.com.sg) Received: from blanc.magix.com.sg (blanc.magix.com.sg [202.166.43.1]) i536i78Z016809; Thu, 3 Jun 2004 14:44:07 +0800 Received: (from shanali@localhost) by blanc.magix.com.sg (8.12.10/8.12.10/Submit) id i536hwrk038715; Thu, 3 Jun 2004 14:43:58 +0800 (SGT) (envelope-from shanali) Date: Thu, 3 Jun 2004 14:43:58 +0800 From: S H A N To: Tom Arnold Message-ID: <20040603064358.GA38572@blanc.magix.com.sg> References: <20040601230306.GN9728@moo.sysabend.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20040601230306.GN9728@moo.sysabend.org> User-Agent: Mutt/1.4.2.1i cc: performance@freebsd.org Subject: Re: Mysql Performance issues on BSD4.9 X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 03 Jun 2004 06:44:11 -0000 hi, sorry for not directly replying to your question but just to clarify i have mysql-4.20 replicated running on its new pthread (FreeBSD 5.x native threading) support on a Raid-5 configuration on two dual 1.3ghz Xeon Dell 2550 without any performance problems :) S H A N On Wed Jun 02, 2004 at 07:03:06AM SGT, Tom Arnold wrote: > I'll try to give background and be as brief as possible. > Hardware : Dell 2650, dual 2.4ghz Xeon, 3gigs RAM. > > Original configuration > BSD4.8 > Raid 1+0 ( or is it 0+1 ) using onboard Perc3 controller. ( 4 70gig drives ) > plus an additional 70gig drive in the 5th slot. Raid was full so we added > the 5th drive and expanded the Mysql database onto it. Speed was > "tolerable" > > New configuration > BSD4.9 > 5x140gig drives ( 15k LVD ) on same Perc3 controller. > > Raid5 - Way incredably painfully slow. Slave DB would take days to catch up > with Master. > > Raid0+1( or is it 1+0 damn adaptec ) + 3 seperate disks. All on Perc3 > controller. - Still slow. We've used the pair of mirrored disks as the boot > disks and the mysql DB is broken up across the 3 seperates. The machine is > still sluggish. I'll paste a systat -v at the end of the message. > > I'm going to apologise right off with not being more detailed about what I > mean by "slow" and "sluggish". If you can tell me more detail to look for > I'll be happy to supply. > > Now, my theories for you to poke holes in. > 1. The original configuration was speed-tolerable because it was basically > one large logical drive ( ignoring the standalone drive we added ) and it > was smaller drives. > 2. The RAID5 configuration was dirt-slow because its Raid5 and you take a > large performance hit ( 4X normal write time? ). > 3. The RAID10+3XStandalone is slow because now we have 4 seperate drives ( 1 > logical 3 physical ) going via the same controller. > 4. Optimal speed for this would be, say, 14 smaller ( 36 or 70gig ) drives > in RAID10. This is kind of my silver bullet solution based on experiences > running INND all those years ago. > > Here's a Systat -v from the Raid10 + 3 seperate drive configuration. > Thanks! > > Mem:KB REAL VIRTUAL VN PAGER SWAP PAGER > Tot Share Tot Share Free in out in out > Act 1222516 2156 1303312 3476 109292 count > All 3081632 4680 3870300 8488 pages > 2085 zfod Interrupts > Proc:r p d s w Csw Trp Sys Int Sof Flt cow 3824 total > 20 2105 12916 382639428 3824 5984 2109 252952 wire ata0 > irq14 > 1239708 act 2881 bge0 > irq11 > 15.8%Sys 1.5%Intr 1.2%User 19.6%Nice 61.9%Idl 1484452 inact 486 aac0 > irq16 > | | | | | | | | | | 104520 cache fdc0 > irq6 > ========+---------- 4772 free atkbd0 > irq > daefr 201 clk > irq0 > Namei Name-cache Dir-cache prcfr 256 rtc > irq8 > Calls hits % hits % react > 72 72 100 pdwake > pdpgs > Disks aacd0 aacd1 aacd2 aacd3 acd0 fd0 md0 intrn > KB/t 59.64 64.00 64.00 64.00 0.00 0.00 0.00 197120 buf > tps 11 97 76 61 0 0 0 145 dirtybuf > MB/s 0.63 6.05 4.76 3.79 0.00 0.00 0.00 199064 desiredvnodes > % busy 1 56 57 39 0 0 0 49766 numvnodes > 37974 freevnodes > > -- > ------------------------------------------------------------------------ > - Tom Arnold - When I was small, I was in love, - > - Sysabend - In love with everything. - > - CareTaker - And now there's only you... - > -------------- -- Thomas Dolby, "Cloudburst At Shingle Street" - > > _______________________________________________ > freebsd-performance@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-performance > To unsubscribe, send any mail to "freebsd-performance-unsubscribe@freebsd.org" From owner-freebsd-performance@FreeBSD.ORG Fri Jun 4 15:05:55 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E3D8616A4EF; Fri, 4 Jun 2004 15:05:55 -0700 (PDT) Received: from shadow.wixb.com (shadow.wixb.com [65.43.82.173]) by mx1.FreeBSD.org (Postfix) with ESMTP id 69BAE43D39; Fri, 4 Jun 2004 15:05:55 -0700 (PDT) (envelope-from jbronson@wixb.com) Received: from thinkpad.wixb.com (thinkpad.wixb.com [10.43.82.5]) i54M5ZsH000664; Fri, 4 Jun 2004 17:05:35 -0500 (CDT) Organization: Aurora Health Care, Milwaukee WI USA Message-Id: <6.1.1.1.2.20040604170255.00bf3b38@cheyenne.wixb.com> Date: Fri, 04 Jun 2004 17:05:34 -0500 To: freebsd-questions@freebsd.org From: "J.D. Bronson" Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed X-Miltered: at shadow.wixb.com with ID 40C0F22F.000 by j-chkmail cc: freebsd-performance@freebsd.org Subject: more on fsck with securelevel X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 04 Jun 2004 22:05:56 -0000 I did set this in /etc/rc.conf: fsck_y_enable="YES" But I was wondering if this might be a good idea too: (looking at the defaults) fsck_y_enable="NO" # Set to YES to do fsck -y if the initial preen fails. background_fsck="YES" # Attempt to run fsck in the background where possible. background_fsck_delay="60" # Time to wait (seconds) before starting the fsck. ..might it not be prudent to set 'background_fsck="NO"' when running in secure mode? Eventhough I shut down carefully, sometimes it still feels the need to run fsck (even with soft updates)...but when running securelevel, is it actually going to accomplish anything? -jeff From owner-freebsd-performance@FreeBSD.ORG Fri Jun 4 15:22:44 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0977316A4CE; Fri, 4 Jun 2004 15:22:44 -0700 (PDT) Received: from odin.ac.hmc.edu (Odin.AC.HMC.Edu [134.173.32.75]) by mx1.FreeBSD.org (Postfix) with ESMTP id B35FE43D49; Fri, 4 Jun 2004 15:22:43 -0700 (PDT) (envelope-from brdavis@odin.ac.hmc.edu) Received: from odin.ac.hmc.edu (IDENT:brdavis@localhost.localdomain [127.0.0.1]) by odin.ac.hmc.edu (8.12.10/8.12.10) with ESMTP id i54MMWSK027274; Fri, 4 Jun 2004 15:22:32 -0700 Received: (from brdavis@localhost) by odin.ac.hmc.edu (8.12.10/8.12.3/Submit) id i54MMWqE027271; Fri, 4 Jun 2004 15:22:32 -0700 Date: Fri, 4 Jun 2004 15:22:32 -0700 From: Brooks Davis To: "J.D. Bronson" Message-ID: <20040604222232.GB25234@Odin.AC.HMC.Edu> References: <6.1.1.1.2.20040604170255.00bf3b38@cheyenne.wixb.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="eJnRUKwClWJh1Khz" Content-Disposition: inline In-Reply-To: <6.1.1.1.2.20040604170255.00bf3b38@cheyenne.wixb.com> User-Agent: Mutt/1.5.4i cc: freebsd-performance@freebsd.org cc: freebsd-questions@freebsd.org Subject: Re: more on fsck with securelevel X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 04 Jun 2004 22:22:44 -0000 --eJnRUKwClWJh1Khz Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Fri, Jun 04, 2004 at 05:05:34PM -0500, J.D. Bronson wrote: > I did set this in /etc/rc.conf: > fsck_y_enable=3D"YES" >=20 > But I was wondering if this might be a good idea too: > (looking at the defaults) >=20 > fsck_y_enable=3D"NO" # Set to YES to do fsck -y if the initial preen= =20 > fails. > background_fsck=3D"YES" # Attempt to run fsck in the background where= =20 > possible. > background_fsck_delay=3D"60" # Time to wait (seconds) before starting the= =20 > fsck. >=20 > ..might it not be prudent to set 'background_fsck=3D"NO"' when running in= =20 > secure mode? >=20 > Eventhough I shut down carefully, sometimes it still feels the need to ru= n=20 > fsck (even with soft updates)...but when running securelevel, is it=20 > actually going to accomplish anything? I think just setting background_fsck_delay=3D0 may allow bgfsck to work. Once fsck has opened the FS, I think it should keep it open and writes should work. I'm not 100% sure of that though. -- Brooks --=20 Any statement of the form "X is the one, true Y" is FALSE. PGP fingerprint 655D 519C 26A7 82E7 2529 9BF0 5D8E 8BE9 F238 1AD4 --eJnRUKwClWJh1Khz Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) iD8DBQFAwPYnXY6L6fI4GtQRAt16AKC5TTVlz+oklnD4Mk2YIKe5b2T4wgCgsuZY Nf08dEIFqZfEwTlI/y/uwjI= =WFkV -----END PGP SIGNATURE----- --eJnRUKwClWJh1Khz-- From owner-freebsd-performance@FreeBSD.ORG Fri Jun 4 16:36:33 2004 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E288616A4CE; Fri, 4 Jun 2004 16:36:33 -0700 (PDT) Received: from internet.potentialtech.com (h-66-167-251-6.phlapafg.covad.net [66.167.251.6]) by mx1.FreeBSD.org (Postfix) with ESMTP id 85D6243D31; Fri, 4 Jun 2004 16:36:33 -0700 (PDT) (envelope-from wmoran@potentialtech.com) Received: from working.potentialtech.com (pa-plum1c-102.pit.adelphia.net [24.53.179.102]) by internet.potentialtech.com (Postfix) with ESMTP id 00B0269A71; Fri, 4 Jun 2004 19:35:48 -0400 (EDT) Date: Fri, 4 Jun 2004 19:35:47 -0400 From: Bill Moran To: Brooks Davis Message-Id: <20040604193547.6d69fa7f.wmoran@potentialtech.com> In-Reply-To: <20040604222232.GB25234@Odin.AC.HMC.Edu> References: <6.1.1.1.2.20040604170255.00bf3b38@cheyenne.wixb.com> <20040604222232.GB25234@Odin.AC.HMC.Edu> Organization: Potential Technologies X-Mailer: Sylpheed version 0.9.10 (GTK+ 1.2.10; i386-portbld-freebsd4.9) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit cc: freebsd-performance@freebsd.org cc: freebsd-questions@freebsd.org cc: jbronson@wixb.com Subject: Re: more on fsck with securelevel X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 04 Jun 2004 23:36:34 -0000 Brooks Davis wrote: > On Fri, Jun 04, 2004 at 05:05:34PM -0500, J.D. Bronson wrote: > > I did set this in /etc/rc.conf: > > fsck_y_enable="YES" > > > > But I was wondering if this might be a good idea too: > > (looking at the defaults) > > > > fsck_y_enable="NO" # Set to YES to do fsck -y if the initial preen > > fails. > > background_fsck="YES" # Attempt to run fsck in the background where > > possible. > > background_fsck_delay="60" # Time to wait (seconds) before starting the > > fsck. > > > > ..might it not be prudent to set 'background_fsck="NO"' when running in > > secure mode? > > > > Eventhough I shut down carefully, sometimes it still feels the need to run > > fsck (even with soft updates)...but when running securelevel, is it > > actually going to accomplish anything? > > I think just setting background_fsck_delay=0 may allow bgfsck to work. > Once fsck has opened the FS, I think it should keep it open and writes > should work. I'm not 100% sure of that though. fsck_y_enable determines what happens when a normal fsck fails. If it's set to "NO", you're dumped into single-user mode with a scary message. If it's set to "YES", then fsck is rerun with the -y option. Setting it to "YES" will allow the system to boot automatically under almost all conditions, but you may lose data that you could have recovered if you'd run fsck manually ... assuming you know what you're doing well enough to recover that data. background_fsck controls whether the initial boot fsck is run in the background, after booting, or in the foreground during the boot, forcing the boot process to wait on it (background_fsck="NO" is basically the same behaviour as pre-5 systems) background_fsck_delay is pretty obvious. Hopefully, that will help you reach a config that works for you. -- Bill Moran Potential Technologies http://www.potentialtech.com