X-Original-To: pgsql-hackers-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.144]) by svr1.postgresql.org (Postfix) with ESMTP id 5214B3A421A for ; Mon, 25 Oct 2004 22:31:02 +0100 (BST) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) with ESMTP id 89626-07 for ; Mon, 25 Oct 2004 21:30:56 +0000 (GMT) Received: from stark.xeocode.com (gsstark.mtl.istop.com [66.11.160.162]) by svr1.postgresql.org (Postfix) with ESMTP id D5DD63A4211 for ; Mon, 25 Oct 2004 22:30:56 +0100 (BST) Received: from localhost ([127.0.0.1] helo=stark.xeocode.com) by stark.xeocode.com with smtp (Exim 3.36 #1 (Debian)) id 1CMCQX-0004qr-00; Mon, 25 Oct 2004 17:30:53 -0400 To: Greg Stark Cc: pgsql-hackers@postgresql.org Subject: Re: [PATCHES] ARC Memory Usage analysis References: <1098471059.20926.49.camel@localhost.localdomain> <41796115.2020501@Yahoo.com> <20041022200955.GC4382@it.is.rice.edu> <417D1D01.3090401@Yahoo.com> <20651.1098720192@sss.pgh.pa.us> <87mzyavhxc.fsf@stark.xeocode.com> In-Reply-To: <87mzyavhxc.fsf@stark.xeocode.com> From: Greg Stark Organization: The Emacs Conspiracy; member since 1992 Date: 25 Oct 2004 17:30:52 -0400 Message-ID: <87hdoiv62b.fsf@stark.xeocode.com> Lines: 51 User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.3 MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="=BATF-Ruby-Ridge-Europol-USDOJ-Vince-Foster-FIPS140-Albright-Baranyi-" X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=0.0 tagged_above=0.0 required=5.0 tests= X-Spam-Level: X-Archive-Number: 200410/850 X-Sequence-Number: 60353 --=BATF-Ruby-Ridge-Europol-USDOJ-Vince-Foster-FIPS140-Albright-Baranyi- Greg Stark writes: > However I wonder about another approach entirely. If postgres timed how long > reads took it shouldn't find it very hard to distinguish between a cached > buffer being copied and an actual i/o operation. It should be able to track > the percentage of time that buffers requested are in the kernel's cache and > use that directly instead of the estimated cache size. I tested this with a program that times seeking to random locations in a file. It's pretty easy to spot the break point. There are very few fetches that take between 50us and 1700us, probably they come from the drive's onboard cache. The 1700us bound probably would be lower for high end server equipment with 10k RPM drives and RAID arrays. But I doubt it will ever come close to the 100us edge, not without features like cache ram that Postgres would be better off considering to be part of "effective_cache" anyways. So I would suggest using something like 100us as the threshold for determining whether a buffer fetch came from cache. Here are two graphs, one showing a nice curve showing how disk seek times are distributed. It's neat to look at for that alone: --=BATF-Ruby-Ridge-Europol-USDOJ-Vince-Foster-FIPS140-Albright-Baranyi- Content-Type: image/png Content-Disposition: inline; filename=plot1.png Content-Transfer-Encoding: base64 iVBORw0KGgoAAAANSUhEUgAAAoAAAAHgCAMAAAACDyzWAAABKVBMVEX///8A AACgoKD/AAAAwAAAgP/AAP8A7u7AQADu7gAgIMD/wCAAgECggP+AQAD/gP8A wGAAwMAAYIDAYIAAgABA/4AwYICAYABAQEBAgAAAAICAYBCAYGCAYIAAAMAA AP8AYADjsMBAwIBgoMBgwABgwKCAAACAAIBgIIBgYGAgICAgQEAgQIBggCBg gGBggICAgEAggCCAgICgoKCg0ODAICAAgIDAYACAwODAYMDAgADAgGD/QAD/ QECAwP//gGD/gIDAoADAwMDA/8D/AAD/AP//gKDAwKD/YGAA/wD/gAD/oACA 4OCg4OCg/yDAAADAAMCgICCgIP+AIACAICCAQCCAQICAYMCAYP+AgADAwAD/ gED/oED/oGD/oHD/wMD//wD//4D//8BUJrxzAAAKoklEQVR4nO3djY7iNgCF UVDe/53bMuG3JMTGyY3DOdKKroRJ2PlqDxkGn04AAAAAAAA/5Xzxd3uauIHV jIX9Nfj+BtZzmwAFSMK9tfPUze2esFRphOeZCK/3qk1802FOs+m4bQ43tfY+ L8GHfOpfj+vkNPf79ObKE+DuDnfApzcu2HM3m51Lg2FOs+m4rU9z68fkoATI g9tCNvfqdPxxxNtx5UesHLf1Y7KJpwCnohq/6Rrvf34a98Ywf8SKs/xEgN16 mgAnoromd3rIdG4CFCBL3Tq6/tD/8SLHdZZ7E+DsAixASj00dn0Tyutk99Tq tOHPzKEanO4Wj8mGXqa6pwBfXqUsePVhBqTQ/9bayQCXvPgVIGUeGnv8Ydf5 6bu9j689bgRImYfIzk8z4Olx5T1fq/zuyy1AFlnr7Z4CZBEBckgCJEqARAmQ KAESJUCiBEiUAIkSIFECJEqARAmQKAGSMwzDefYN+3UEyGJmQKIESJQAiRIg UQIkSoBECZAoARIlQKIESFRZLA+fxrDCh5Tzg4piOT9/RkjrbRr4QSWx3D+P RoA0UhDLy6d1XefDh5vbHb/9yCR+QlkoD7Pc2/rMgFQoCfDxY6otwTRR8SpY gLRTHOD76y8uw1DHhWiiBEiUAIkSIFECJEqARAmQKAESJUCiBEiUAIkSIFEC JEqARAmQKAESJUCiBEiUAIkSIFECJEqARAmQKAESJUCiBEiUAIkSIFECJEqA RAmQKAGSMwzD+d8/rR9WgCxmBiRKgEQJkCgfUk5U1V5xtmmglYq94gRIOzV7 xb3dpstecZSq2itusj4zIBWK94qzBNOSveKIchmGKBeiiRIgUQIkSoBECZAo ARIlQKIESJQAiRIgUQIkSoBECZAoARIlQKIESJQAiRIgUQIkSoBECZAoARIl QKIESJQAiRIgUQIkxzYNpJkBiRIgUQIkSoBECZAoARIlQKIESFTRRjXXzyif /Kzy4sfk11VsVPP6N9s0UK9in5DzNTcB8rWKJfi+a5y94vhGeSj39ddecTRR H6AlmAaqXoQIkCbK3g84t0mcyzDUcSGaKAESJUCiBEiUAIkSIFECJEqARAmQ KAESJUCiBEiUAIkSIFECJEqARAmQKAESJUCiBEiUAIkSIFECJEqARAmQKAES JUCiBEiUAIkSIFECJEqARAmQKAESJUCiBEhU8UY1PqSclmzTQFTFXnECpJ2a veLebtNlrzhKVe0VN1mfGZAKlmCivAghymUYolyIJkqARAmQKAESJUCiBEiU AIkSIFECJEqARAmQKAESJUCiBEiUAIkSIFECJEqARAmQKAESJUCiBEiUAIkS IFECJEqARAmQKAESJUCiBEiUAMkZhuH875/WDytAFjMDEiVAogRIlACJqtio 5n57er0pfkx+XdU2Dc9/s00D9SoCvO7YJUC+98VOSfaK4zulodzuba84Wil6 EfLwX5Zgvlf0o7i5Vx8CpFLJi5CZ6y8uw1DHhWhyvBuGsMEMSJIACbIEkzQI kDBLMFECJGdcgps/rgBZygxIlACJEiBRAiRKgEQJkByXYUgzAxIlQKIESJYA iRIgUQIkSoBECZAoAZLkMgxRAiRKgOR4MwJpZkCiBEiUAMkSIFECJEqARAmQ qOK94nxIOQ0VvAqe25/BNg3UESBRhdcBpzaJs1ccFUp/Fnzdp8ZecTRRGOBf YZZgGir8HvAkQJoqCtBecTTl/YCkeTcMUQIkxxJMlABJswQTJUByLMFECZA0 SzBRAiRKgGQJkCgBkmQJJkqARAmQHBeiSTMDEiVAogRIlACJEiA5XgUTJUDS LMFECZAoARIlQKIESJQAyXEZhjQzIDlmQMIGMyA5ZkCShvJ9Qk4fPqt8vF/z U+Wgipbga2rjX2zTwHfGGXDp3c8PM6AAaaLsRchtBrRXHC2cz0NRKLc72yuO BgqX4NcALcF8py5Ae8XRzlAQi73iaKt4BlxKgCwhQKIESJQASRoESNIY4AqP LEA+GwRIzjAIkJhbfiUXopcTIHOGh/4EyKaGJycBsqFh+F9/AmQjr/WNvwsi QDbw//quv4skQNb2rr7bzz8EyIretjc8/iamAFnJVHzPvwgsQFawLL7/CJDG puN798YXAdLOTHtTn8AhQNqYjW/6TX8C5Gvz7c1/+pAA+cY37V0IkEqf2lv2 XnsBUqFNfP8RIGU+tlf2W0YCZLHG7V0IkCU+t1f565UC5IPV2rsQINPWbe9C gLyzIL02H2kgQF5s1t6FALlZkl7rfWUEyH8S7V0I8NctS2+N9i4E+LMWlrde excC/EGL01u3vYvyfUJ8SHm/lpe3QXp/qvaKs01DdwrS26y9i4JtGuwT0qOS 8rZN70/NXnFvt+myV9zeFKWXaK84lOt3e/aK27ey8iLp3X2xW6YleG+6Km8k wEPosLyRveL61uOk98SF6E4VlrfD9P4IsDdHKW8kwH4cLL0/AuzBEcsbCXDf jlveSIA7deBJ74kAd+dHyhsJcE9+qbyRAHfi58obCTDt9ya9JwKM+cH19g0B BkjvToCbUt4rAW5Eeu8JcG0l6+1vtXchwPUUpfeD7V0IcA3KW0yAbSmvkAAb MenVEWAD0qsnwG8o72sCrCS9NgRYTHotCbCA9NoT4BLLy5NeIQHOU97KBDih YNIT3xcE+Ep5mxLgXVF62mtDgKff+R3cPfr1AKUX9rsBKm8XfjJA6e3HjwWo vL35nQCVt0u/EaD0dqs4ltnPKq98zFUpb9cqArzd7H+bBu3tXt0M2EWA0utB 1QzYwV5x2utAdSh73ytOev2oD3C3S7D2evLFErzLALXXmfJY9rxXnPq6c6AL 0eLr0WECVF+fDhKg+np1iADV168jBKi+jvUfoPy61n2A8utb7wGqr3OdByi/ 3nUdoPz613OA8juAjgPU3xH0G6D8DqHbAPV3DL0GqL+D6D3ADQ7FmjoNUH9H 0WeA+juMrgNc/TisrssATYDH0XOAax+GDQiQqB4DtAIfSMcBrnwUNiFAonoM 8PRfg+sfgy30GuD6h2ATAiRKgEQJkKgeA/Qq+EB6DPBkBjyOPgM0/R1GlwFa f4+jxwB9D3ggewpw4bjX9yKsfLhW4zo5zV6e3uvDPH1I+bCFr5/Cwb9CB396 L4/yvE3Dpv35Cu1i3K8F2OApHPwrdPCn9/Io973i/o3jDIs1CvBxr7hN5z94 XYL1x8YESFazxRwAAAA6U/ia+O/O46Dnm88HKR1WPe50vdpUNO56tX+r00wc r3jYzBe8yeWU56uCC+5+vg96vvl8kNJh1eNO95/3lJ5mxbP75jS3/GepGjbz BS9MZ/6slt77tGmA49iKcee6AM/XOxUHMfUsPz+7XQc49wVvFmDRRHo9n/EL /HCz5EClw25FFI+rOd79H7bscOfa03w45Ab/LHVPb/oLXprO5HMpyng8nzcn 8/EwFcNO08/900nWjNv8cHX/mrXHO4/rQtGw6VMsTWf2yRTd/+H/pYebj6PK h42DKg5XtQTXH67+O4Wp4escr2LY9CmWpjP7ZIruX3wi58pvkr553hXjag9X fZqV/19WHq9u2NoBnopeS1deOKgcNjFg2RnXHK/2cPXjpoevcLyKYbNfuaJ0 AAAAAACgzj9lVEYweskfcwAAAABJRU5ErkJggg== --=BATF-Ruby-Ridge-Europol-USDOJ-Vince-Foster-FIPS140-Albright-Baranyi- This is the 1000 fastest data points zoomed to the range under 1800us: --=BATF-Ruby-Ridge-Europol-USDOJ-Vince-Foster-FIPS140-Albright-Baranyi- Content-Type: image/png Content-Disposition: inline; filename=plot2.png Content-Transfer-Encoding: base64 iVBORw0KGgoAAAANSUhEUgAAAoAAAAHgCAMAAAACDyzWAAABKVBMVEX///8A AACgoKD/AAAAwAAAgP/AAP8A7u7AQADu7gAgIMD/wCAAgECggP+AQAD/gP8A wGAAwMAAYIDAYIAAgABA/4AwYICAYABAQEBAgAAAAICAYBCAYGCAYIAAAMAA AP8AYADjsMBAwIBgoMBgwABgwKCAAACAAIBgIIBgYGAgICAgQEAgQIBggCBg gGBggICAgEAggCCAgICgoKCg0ODAICAAgIDAYACAwODAYMDAgADAgGD/QAD/ QECAwP//gGD/gIDAoADAwMDA/8D/AAD/AP//gKDAwKD/YGAA/wD/gAD/oACA 4OCg4OCg/yDAAADAAMCgICCgIP+AIACAICCAQCCAQICAYMCAYP+AgADAwAD/ gED/oED/oGD/oHD/wMD//wD//4D//8BUJrxzAAAILklEQVR4nO3di1baSgBA UbL8/3+uiiCoWJMZPTz2XsXcW0sY09OZJNjV3Q4AAAAAAODBLMv+4+v2fAO/ 7lDc6+N8A79uX5sAyRwCXPblnWyOvwBWWh/gx/pOApxY+zCDueRmB/PfJfhm v7JfZzCXCPBPGMwlPx/MYcV+W7fPN2t39gcM5pL7HcxVfWXcBAFy4riefXd1 +rYOfvm89a+48Xm/vzMCZwFeiup4LfB+BvZdgE/fv+KGUf7NzgicTYAXojok tzvJ9LsJUID81LGjw9Xm6b2Owyz3RYDfLsACZK2Txg53Pz5OdmetXva0981L TRju7+yMzIep7izAD1cpP7j6MAOy0qe19mKAP7n4FSDrnDR2+p7Xcna2999r jyMBss5JZMvZDLg7XXmXQ5Vjv+sC5Ed+67s+BciPCJC7JEBSAiQlQFICJCVA UgIkJUBSAiQlQFICJCVAUgIkJUBSAiQlQFICJCVAUgIkJUBSAiQlQFICJCVA UgIkJUBSAiQlQFICJCVAUgIkJUBSAiQlQFICJCVAUgIkJUBSAiQlQFICJCVA UgIkJUBSAiQlQFICJCVAUgIkJUBSAiQlQFICJCVAUgIkJUBSAiQlQFICJCVA UgIkJUBSAiQlQFICJCVAUgIkJUBS65tZnn3abN0Zj251My+9PT/ON1t3xsMT IKmtS/BrfMthc/IpEfJTm3r5XJ8ZkO0swaQESMptGFJuRJMSICkBkhIgKQGS EiApAZISICkBkhIgKQGSEiApAZISICkBkhIgKQGSEiApAZISICkBkhIgKQGS EiApAZISICkBkhIgKQGSEiApAZISICkBkhIgKQGSEiApAZISICkBkhIgKQGS EiApAZISICkBkhIgKQGSEiApAZISICkBkhIgKQGSEiApAZISICkBkhIgKQGS EiApAZISICkBkhIgKQGSEiApAZISICkBkhIgKQGSEiApAZISICkBkhIgKQGS EiApAZISIKn1zSzPPm227oxHt7qZl96eH+ebrTvj4W0KcPexw6074+FtCHA5 xrccNuefgx/Z1Mvn+syAbOcckJQASbkNQ8qNaFICJCVAUgIkJUBSAiQlQFIC JCVAUgIkJUBSAiQlQFICJCVAUgIkJUBSAiQlQFICJCVAUgIkJUBSAiQlQFIC JCVAUgIkJUBGPY08WYCMEiApAdJ52tv8fAEyygxISoCkBMjtEiApATJoaAUW IKMESEqAdEbvQwuQUWZAUgIkJUBumAAZZAYkJUBSAqTjPiA1MyApAVIaWX93 AmTUWH8CZJAA6QxfBAuQQWZAUgIkJUBSAiQlQFICpOM2DLHBd+IEyBgBErIE UxrvT4AMGcxPgIwRICFLMKUJ/QmQ7QRIbfQUUIBsZwakNjoBCpAhAiQlQErD 1yACZDsXIbSG34gTINvNmAC3NbMsL4+Xj4fNwM64UU9ZgC/N7Rs8bAZ2xq2a 0t+WZhYBsjd+CrihmX12rz+Om+OnThdk7t7oFfCWXo5T32mEh88NDYebMz4B rm9mWSzB7E1YgTdfBQuQNEC3YR5edhX8Rzvjus3pT4BsJEBSAiQ24xpEgGwz aQIUINsIkJQAKc35XqydANlqRn07AbKVAEkJkM60U0ABstGcCVCAbCRAUgIk Nac/AbKRAOlMuwgWIBuZAUkJkJQASQmQjosQamZASpPeCBEgW1iCiZkB6cyb AAXIBgKkNmkFFiDbCJCUAEkJkJQA6bgKJjbrPrQA2USApARIZ+IpoABZT4DU Zq3AAmQTAZISIKVp/QmQLQRISoB0Zt6FESCrCZDYtDfiBMh6ZkBiZkA6UydA AbKBGZDUvP4EyAYC5F4IkJQASQmQlABJCZCUAEkJkJQAWc+NaFICJCVAOr4d i5oZkJQASQmQeyFAUgIkJUDWcw5ISoCUJv6tTAGylndCiJkB6cydAAXIeu0M uDz7tNm6M27RxP7WN/PS2/PjfLN1Z9yeegkW4IOrA3x90rL/cdwcf/50QeZe TbsFs62XZT/tnUZ4+MykgXHd0nPAt94swQ8svwjZCfChxQEuhzXYbZhHdA0X IX+yM67VzPvQAmQ1AdKZvAILkNVmToACZDUBkhIgHeeA1MyAdMyA1MyApARI SoDcDwGSEiApAZISICkBkhIgKQGSEiApAZISICkBkhIgKQGSEiApAZISICkB khIgKQGSEiApAZISICkBkhIgKQGSEiApAZISICkBkhIgKQGSEiApAZISICkB khIgKQGSEiApAbLS1H8mRICsJUBSAqQz+x/LFCBrmQFJCZCUALkjAiQlQFIC ZCXngKQESEqAdLwTQs0MSEqApARISoCUZl6B7ATIOq6CSQmQ0vT+BMjPPT0J kMrTiYm7FSAXPF0081UEeLUuB1Cb+VVeW4D1seW/xn+TT80NsD42/L6ZwewE yFozg9mNB7g8e/+f+uDwqwZb+Tqgwacv+8fu9exNgPdsQm1fFTT49PcAd2bA OzPa1s8KGnz6sjtZgxdYazTA5WQGvJk/dtyL8yUY/pgAaY2v4gAAAHCbruKa +HCD/W0w6Zj2L30+kmxAx1e/gqPz5SEZHsx13BV8G8HbYNIxHY7r6UiyAb0P 5tOYgrF8cUjGB3MtAS6HgcQBLrsrCvA4mGs4Or8X4BWswe9fzvK+CcfyYSTd gD78aaiPzqdRjA/m/FsTSp++sm4cuy+OcRjg239cwWCW+UfmOpbgF/2KdxzH tSzBHwOsT0gPH+/xHPAqDvFxLFczoCsazPLVpdCEwVzFbZjDKOq7HpdueUQD MhgAAAAAAHjzD9bGNv6j2lw6AAAAAElFTkSuQmCC --=BATF-Ruby-Ridge-Europol-USDOJ-Vince-Foster-FIPS140-Albright-Baranyi- This is the program I wrote to test this: --=BATF-Ruby-Ridge-Europol-USDOJ-Vince-Foster-FIPS140-Albright-Baranyi- Content-Type: text/x-csrc Content-Disposition: attachment; filename=timing.c #include #include #include #include #include #include #include #include int main(int argc, char *argv[]) { int rep = atoi(argv[1]); int i; char *filename; int fd; struct stat statbuf; off_t filesize; unsigned blocksize; void *blockbuf; filename = argv[2]; fd = open(filename, O_RDONLY); fstat(fd, &statbuf); filesize = statbuf.st_size; blocksize = statbuf.st_blksize; blockbuf = malloc(blocksize); srandom(getpid()^clock()); for (i=0;i /tmp/l $ gnuplot gnuplot> set terminal png gnuplot> set output "/tmp/plot1.png" gnuplot> plot '/tmp/l2' with points pointtype 1 pointsize 1 gnuplot> set output "/tmp/plot2.png" gnuplot> plot [0:2000] [0:1000] '/tmp/l2' with points pointtype 1 pointsize 1 -- greg --=BATF-Ruby-Ridge-Europol-USDOJ-Vince-Foster-FIPS140-Albright-Baranyi---