From 6f76dade24d7ba68398c46f84c1bd5c1d6bb7001 Mon Sep 17 00:00:00 2001 From: "Galen M. Shipman" Date: Sat, 16 Mar 2024 20:03:42 -0700 Subject: [PATCH] update parthenon perf numbers and SSNI baseline draft, other minor fixes --- doc/sphinx/00_intro/SSNI-baseline-draft.xlsx | Bin 12458 -> 12291 bytes doc/sphinx/03_vibe/cpu_20.csv | 6 +++--- doc/sphinx/03_vibe/cpu_40.csv | 10 +++++----- doc/sphinx/03_vibe/cpu_60.csv | 10 +++++----- doc/sphinx/03_vibe/gpu.csv | 10 +++++----- doc/sphinx/03_vibe/vibe.rst | 19 +++++++++---------- 6 files changed, 27 insertions(+), 28 deletions(-) diff --git a/doc/sphinx/00_intro/SSNI-baseline-draft.xlsx b/doc/sphinx/00_intro/SSNI-baseline-draft.xlsx index a0fe694a3cc248c4ae88df2ecea2cfdd4955bf68..700de4cc2249951516fff3752b553ece9f72d28b 100644 GIT binary patch delta 5642 zcmZ8lWmME(*Bv@W8tLxtPU)dTKw7%H`$rCqzyJ!PbcoU*4Z?twbR!)Kh=g?W@_(MU z?}vNVI%}`<;heMYx_f`Jingq!!~%8(`pWW9K%g5;5C{(h0tLGA26%Y8SbKQ5@C3TK zRvPGf6pP`9%kDj(Y_15iyZ{NC%hCOSjJjvpFLjm=Be5Exy%#J~H8=pi@?$x*i&uHC zBoRtB#f`6xgbJF4|1c>1+h^OWBrPx$Ye!KzrZ~2LK6uykj&4neLTUefHc({Gc$y>p ztNgRvqx0(jiKF6w}HM)6J=(CGYY=Ch4=Y# zWUg{n?6e;@lX>>~MQTgG7J!(KrmeHAy@jvuZhuO2&=ayQ5=q_r}A6=8-N` zH=$Z7mRi@BC8j72U*o0gUKwnZq#q zP;=00cuN-#iIa;3IA52=@Q5ki1jI5q|#HiKL@Be!8c@b@1MK1mA@xb&;A&K{E z&^=1Bc@S1}pCr~dEn0$$-%yPgeBIBkM8TETgx)BNkG6m6mWC*Qf2O9OZElau_Wp%R z#KDYTaT~e*5D>}bRq5!6hFd=snG}(m66#=CTnAgIkko()q1T6TUeV?{yrQytl$qiJ zbIi(5{UKPj_GwCVLfQNwi5j|=kzy{}M29h$lZNFDu`Cq3pxS1EMM^RaiTSbC*nYYF zjMv0N8=x)t$%M_td~!u%W^21PWZCRM6xTT9RosTuq>PuW)IDdf>2a5>1BLeCm=G0^ z!)@>I!=gMX?sllPf09e9==7lYTI$9nJ^ZMq`;kItnc3{k2MsS?*k=F)vtlQP{p7)i zq3{wnoIk@u26~9n7XqL7ybK0`aQ@9_AKxHXJD-2p++n!rz9dE*Mr-*Hx!zvHoG~E6 z^xTgM2~9#N(l#Qflxdh{SgBxF-<|()Rb96!OK3=JkC0I$NHwpWcMopi>8s*5kU8QR zNMNaGSL)0nO8&6Zbo`6V4?39Z$?OEZZ0v=_;BmjcNM8>pMiqRG&r+vgl=zzpU+%2j zeBVH=v9C76%l-m|M7hYMYY9@SIwUtEldKwh<5|WyvyTFcyrHJ;;i!_BeFyG7H<+$X z4Cg&WyuK=tphr5{Peb(K6@|22-eke>@wqA1cASd0u~?8&s<_|gm)vlHL$+%Vs-zCO z3Sh7B>48-oE5TKwXYa)@x5ZkH2=n-_9?kD)wu(jVph*adk(@sc?3d3P#A@^NorX{v zcHah#&>~|lkBG@aZBREHv4XI3Ts~%_xlXintu*8F5hf=4+<6`W?jAymak>IQ>v+#j z8}-|uMIxH0D#I_eC{c7v7VTYt6z7e0{r(r5JAl*Zpt35$O@zuKISDrk?j3FeUi&&K z{iiF3i7Zp+bLXgVJE~HDX|E4ulB|-?!b>A~2k(GVn2+u8%>J=zMQRO!lWA9Tg$&DI z>B?Oy99qk;g@laX*1PC$th6faH)xxGF7&*521f`TYK2 zw3fVwizQRRaOpf)M?tsrQsAzq3_EU4l0}FkAeoYdwt_ao__91v=nUH?uAtD6x;3Iq z*ljQszO-cp)!zT=H0qBm@78gizQQi}!WOt{4EF##h&3p^z4#S|K?J%h`e0A~5Zrm= zQrE@9;6YRQ%OX^j$RoXS&VulD)!e={fnv`r0Mq@0fnI0j9YaXo|2?(0rA)<6Wy3?O z@tL-W)?J}p#KOx((n+pBcueFgGxS2z@xf*II1f<`1mP&1^lDx;UBgz}LcA6&oQ$q* zavVFK3x_CEFz9%;N8u%h7*^9b+Xk)FMt=Xv9p%4>gDd&QfiB>QC*qiS|vKgcGekWQ(*b+Yj7^?V#Cvum{M`jp@07MQa*D8vZ%<2)JAqI zxwE--y`{@pkFj5}hfSd=uU0@*mc3F|@HMPVd@Id<6w2)r+ zaQp}8SUx=o@#XH?m-+zGT$0Omq2?uXTFl}QMSVS0fTA$dG5xEngiTOs?U7BUw_GYo z4cXB)gJ9|!n-hDbSJ_`8riO-{r3(kw{+5;u_uH>vqS6tB1uo?kG*Q-`J=X#-{!7Vq zO~=)afUen{ey9ZBh$du@?^kj~d$#4Gk2~EEbGLl=4;L z5<)XqzMf|)bmgZ-IWtA{bbe0QnZ?tglPZ8H4yBC`zdPXjixwxZ*ye?Z*hYMp&MeTE z`yL1-e-!-mOjxb{pM zlbH97G7dBfbovL@(zmsDy5@Ob=jCM|<#48T$46 z`z0J_8vYrm2n<{fdpJ5iJGnZ09?pd0A$q&7ayq4QIyZ{mx84K3yJ_1w3BGOnx+;SJ zK-9#AAXq_?oi- zwty+{DQ?-6D{NmZoKxJf^_N-l@hH>-efIuX!Hie#`v%T)H+65I4-raB|O)H*msH|tu=pNY#44+=0my3e&bjdDZ; zu}VxitrGYe93n}Pp;H{_c^9?|cL*TZYuFgVWaEkBf6_==_x#3bU|FV6*lu~wpCRbu zi)?z@R<6gcl`!?=6m#-X$WV?VO3c67v7kS<^AK>rA`;dh8KUar5~b>kp|sDRE9*Km zP*7B7HX~<kd&_RFhY@fM-e^4_3ln*ijUxh0WS=tNCjG3wMNi*=zT}I z=^TW|R>ky6ZB9kOxxja+J{y}o2K{^Fd*G42XDp+SyH=U-M(Yq_Fo zzx&mWF?Euv0F^CtM7=Za^+0S{Z-L??0k6g|VbicbnbWr1#M?i6bT6$FLJq?)83$3Q zGcFr6;z3FnMXl)YzG7W@W5pt-ftbY9ZyNC!fBk(n_;~D)3lv!t9dZS%3>dUhgaB*g z#Md`ZC8BzDL=3E^`U%C1Rr%^&J?RYEU)ZX!l^oLx_PCOX*pOg1VKZljq+LfW+lL<=E#YP0v-{@&OaKpXCN(tYqZvWg36}|Z6Ztb$_!?}mMnqsD?_~@XE0J;JgK_EA*^vK z9)->MC|V-{rP}xBpU9YnIEVTqR3>%wI6X!b7c?U3Eka}XOXTWYT^$#xgxnHGhew26 zqbC?D&?9tbPQ@>lOm31!lChs1kO5I@c^8ufBK(}rEL!&|+0lhAo`g*Z_@_8Y_0(J| zhXCrE7)>x979^VlxAZrwhBs1P(N^q~LXjI`l$I~ow2JRDW2b5Z))Sx{Lb&%N+Ruc6(Ip+skB@s~Dh@3guvq90e} z4~ZYKC2cHVEnoaL>X07UVx#6XjAD^RoZg4+Kh3QU9D4gWIH_|SF$dBcd91KBo~1Y zk%-*Fq@TvO3JutXlIi@2CA%QnDq}`OkTspiJ$K3JF3o&d7~1g~2A2sE+Nv=nVPT$;Vt1yaG7U7SkuaLA z$$6ofA>YwwDJ8fTF}$2R9&eD-+1H<_$2-2Ou;rh!8aXY=`1TPg8k_C8JOW+ZN`;(J z%uE=bbMU6tz&JT8Z!Zlshmh3z0coXD{tXk-kXKRPN8V&K5grNzQSNIgxG0EFhKejG95DwDU{WzZ>v>4A}0(gHqgzN|?f zF(dwM7HYuUuQu)riPDI~I$o9<@`q#u2tIq7Rlb8I=uBC$*92y^AnS_rR4nsLnRNv} zKvwZiXee8t+s0P?LnM-qQyEf@TiUUK0UxLIG~H8+uv-?Rf}{U6F98U@VmjX+%wASc z%~1Eq;z7lM4EYvdgW;A>@5bN1Wpun2#U8)aHl|r;W7o2XDdo-KsJ#6xQ_}>09=vA! zivv2RXMP_81$x%P2H)?et}_j!0>kUQIx!J*1a_Vxu?&hu)&$}l5~T3d;o5K3)4JqM z@ntrtR9nViixZkU>|y_Pcjx8ZYaLV^zvCg^@?&+!tQU0$eTyd+2`V4Y)jHyjvR<6? z2&tRdjR&ZA{F>mg!in4i)0?7=3QwjcLaQadrab=I2L9)7}WD_tFu6ThqM5jd754++Sug*PPi+N@9UeUMa8pGiJ*o`7SlT_4jd~yu#dBrrlZA zF|YDVlIg*)8%5ngNhBb%Diuh=3eQ`!vq_fURz8`dc(;y&8fuWAr;p&bl9meE?^Z!W zV$Q(QXu-1R-zB4$?)>3oGeE0&+ZC?<(3&I(#tgNZ-&e5^Ku6%pJ$zGIc~|= z0WFJiyoN64GL2?Vq?YOuO9aZBx%rpO9XL%1yLC>PpfjrOSKHl_ZVf-_wWc>I(M(T* zUyf%FJGqZdNPF}YI4OmPOfiO)AE?vmstq-L=*oI~^2~Taxp+4 z{C~@J8!J~E1xG6<_kR?-Z`ODBXg++U%d8u`!CUCUSydVr11l`-7Ad|32}>hJclIHy zii4&vxpM@Is`mp1d_{ZBZq45Jz*3$5l}kFgDc;VUh1%b1ufV|PIdG1U>b9O-y*mn4 z|E1e@O>;lsysN?l9u)0Y>R7gSlebgtuJO8c)>C-y@l-fDO!zu~i#cF^Qr}^oOz_;i zdTxKG;}W@UWoFR`n>$}=E?YjK^7IYE38n1VGOgh0%P5y?-_sH33jynRA8y(ldVJ@n z2)4(Jq0qZyE~yK&=MIy3WFPe)#S|WB&F&iP!v`>h`&z?`qd*@CrpVAmjI(_p&2XUL{5FH}$+9Z-_u zorE*&6bEv6y9~=`&bjW0@MH&SeqZ3hCb$={{@YE3)$wQ{EoAVrvi^TV=4r_PdB9J8 z>yzF3Z+iwd$wN%}e+B7(;r@x$urOXFWK$YgAFnL(E;9^`kAdo6G=16~e`FAd;D3Wb b9c(ZKK1Ry_zg+)=(T7FyF{1Nw|C9L-K=@di delta 5770 zcmZ8lbyO7Iww@W5vo#iIHvuhm@3*?x9O!=oAD&T2i`U zhI#z%eed1%-uY+mbEV{>qDY5b2xEBPQ76K;aRm_pJnT zAVwO39J23yPZk&}gFuZ9ORr_Xf6XKv^;iul2|pc3v!*Q@Q0hvb{((mJkN1ceE2Re+ zz;mH`Cu#K?MbUCt@g>5mOXi+DAV5>|?JVA6MAK{t8qv(7`*M|6Dl_EFaYlEG_o?JHthF zM$60y>z(!>bD(|CXi`DTvi%$n>Jn+|1ZVeCK4g7FeC|q-8W&gM!>`1fyhEIG6nCnF z?)e_ls@`=jPEf?sWK_dh;Rv%}S7x$$DrpgG))jF!Z-<*d?a2Od-Jg6!%C&YR`h>l5 zc2BDr-?Xx0+}N;0C^ibo!`Eop`-8q)*J-weWUf3f&d`EASu-LMLh9b0{Guzl4qo~L z6e8Ujyk@sPpLLE44N0=B`5M)mtCQ31Z_-iP#3;p&$F-1J+abkIBl2hqY#Exc0ra=4 zEar*TGSGVfcVvOJPBvHC2ZvYKxebAdO~a+pFY-?tS#N_i>?_`)$kB1j``EEaK$Z}7 zKCC<+h=m|{xlknjL8~8eqx=uxW;=hSE;RXZ+_9f$-7I^(f`sHe$Hm=+DMQx?+0CR$ zIpY{Y1Fms`67ra->h%7!@6XnwiQIu-Md09+juM^!0eR)!(izAA!NVa9>Gbo{Z8=Sv zBPrtbs)@$9?A6w+KfU$VZ?d8nV1KR56c!>6&N2?^bTQ8N*01O6ych(qv>akv6#1{e zrteBz(~m4u{Jf~pk~A?98s{7w%1x1Fly^7&lLMDIn?Cv~<(G+v)Hf{Ov5jXWsOhQy zUj3(~UyXqb_>@l51ofgCo<3VFriBmz)jaROl8IR@0lN+-C!VPDOe;*H@0;m+AY<&n zpm{u8zopNyt*TZM@)IK}VJ53@xT0S0AAcS`>SK%4gZQ$ScAEzr27Fn*wSS>c&{jH+ zG$;5leBKg1^5@-K^Lp<#>fr~O7nrFI_twBIUawpaFW=m^XlrCDSLT*)694`|jb3K= z>%@?voA48#%b$-ivh;!2r(YW(jf}o&tIs&7u(cD!Dj+yhMj#XXj;*4^007Cq+3Mlx z<7Dme4_8|a=HJbWllz-nU;~$1a@E-?omF^t#6kBwqd}^H{snA<9D~vF6U7Ub*tX|7 z$RttKOs85wOT_chw3US^v9{%{xScrew3Wlhoe_g-(;DxD(A&tv4eIw{{a;+!?ZYnW z5GiE*=EJLfM8Y~z9r}U$^7*k2d2J! z1%L6W`#WdP`sFdIaTZAinGVBv%hx6LMoMr4?-hujZe~2-%31}wN-)vZaB&@qU?b z(*7GlY9ggRiGaq7=y^KLb7-OlHTs!FR_4jU!<=sKzbkM(#~9UWm{W;}~ z@$OaQHeWwi)`xkqeiUKs|=TuA&=m7pLhO|N{jBu zgdj#qei4ZVr8fn0h7tGY8FZoyZI^{&7ABIdORFC;@>arleCk=hz8T7FL#!|azD(#o#$j+ zVUk|S#=k{NAts0)-itgmQV_T!3seNrN5KWZW_ofE>4W!_%PNs2Qqol4zaq8?mshk9 zd7?XwT?Fe0v06Wf#oe;}dyot0yN;USH54uHH6UpkVN*_2> zXb|AZ1m|f-e`4BDe(PO#L%x&5^p5?lSn*RTG~#s_dKBb7b8 zcU53Gb)zPaO5q(7G_^1RWboqeQ_ovabxxx)wj!ymWwH*2Th#P>??#^bH*4-^wd)hY z0>x;w^*qfGZ>WEEZX(#d2G$GXG)!|tYpJS3V})b0nbS!=Hzd7V&70!m0E$~hH1cZy z@^C~F*Y3n8R1kM%>!O1n{>^N?KrpR< zA^yYI_cw1_4t+9(;U%92?Vw^wt2KpPa>zGF7(1J8BWqV4%vFs~z$l;+rHt7wI^#51 zF>DAWQ31#lw7ul}TM%NSr?js}uy+Pm9@Xku0DEpvzZ0L0K3JX$9hxXN?N&n}W~q=eOr^fM#< zv=uZZY=5oPbDunFxNW$h|d%SxeW?sARv6TQ3Y z!v?_&w*;t%QQlfS^njXLG!9xD4D}o%-V9NsMiw3JQA#-ux|xcxAxKf{5}yOXd>;Nf z3?iJCzjLLiXH%cGOR#dk;T6?(NbsP}k<})W+~|RS(0*_zY)e4DY%mFSvSRT6xh}OX zbBVv|rYq{>kz6ZQ=qos;;kc|IWX(firs9m}v#GbUdwjmLk34f~Yb%|4uPG|CFB9fv zb%dj#rc!@Mr!-fqo? z!Z;nAhFSVz*>+Tr@et2N9fXIBC5SsB_eUKqb3R-Pyn!uC=>nT#bCJh;EY1DRXeTX_puzmX*%D*nXeWJg9nt>|LP`9;kGGx69sI1^jtG&PS}_B zP#0tVJ>jjF&B*=o(fv+M8K2uA<)rowoGVfCT-BwxZlY=CZxAqAC8Cr*+tbse{T-k$Jodpo&d zY2a_PZL^db@e*q<`fnL;b7>9iE)UIZJFDLH-=h>o2E9O&RttHGeVd=%QrY|QLQ3w% zfuBw5>8Sk5uM|tkDq1FQ%1-cI5QU2sdSb)A>q9tYSuZBMhubb<7sj$~nwGOhV8G1Y z+|)^peodutz-zu;>r+!9Dwl8K`gpAz>5LbQ-LRwwyrMQ!k|ULzpvm>cJjMNo+`pWT z+6CGpM%yxSmMC%l*x6PH;JZS8Q{%Xepl=PSaVEU!aTdk;J{}9^c=Z54>u~C=f*YBr z`N6!_M5~z*2o<__^t=*&Glhz7?>JnIRkxmX@Q8tYEfX}$kBxcG5CvURbBp`>lF)p7g7J#~>bvpg^d8~ks@>+lW%gcti_^Sc2Kx2|;;tW? zX&sx}+EnfCY?MXjFa$L5cfoD*DCspjj6*mB@+kvNsfcYg&3%3Ue0#XO{wgC;U(zA< zhpu0D{tPKCM?LkDK=Jftm?|HN08#qE&rz*yd`|zY8@W@-eq}W&@%Z}*Z?5z+dFuw_ zK##zWDblVF|9%MX?@&m$SpE6%p45VnP zL3>AIW)_}s3BC$Eqxb8k`KVVtNJ_~rSMtDSB}&zqQe|NG^AdE-u3(sfzjFXvISHOqJ-%m ze%oZc!iaOD#EY+WcHrNKh|Llok#_D3jr!HTG)v=!_Ds% z=f3WGi1D$7&{}KxGJ5gI1uf>bD$3zJx*-Rj3;xNURbF~}701olD?2)gC}^4(Lxv;01SvLnPn>IdDmwKDoMrQRo@@w6@OFjW>S>C??0GPhF zHQhTmZqmMchA=WK*q8eQgyk^ia&OnR3raEvV~Z6E_8TRhcOeyOA>d~&1)xbdfm7!^ z#BL$eg`O-Qj`ykE@t6UHv;BRo%tSg~7R~%HG4?Xcm8?bbX!HZ~0%y9Jbm9mM4fy(K zKN248JQ+doGg8|L=O?|(O?<_(hu5Jw4YM30u+Dd_ZC=0oMYmCT2VeY9PKadb(ZTA> zFURMl(ZKn9?j;?czf^V&4NjS7@(p)~Mf$T|m%P)43mC^B`lK62{VYv}4StN&|1Ny0 zJ>GCH07gG0Bf7~VI)|E1uiFIPoD6tth;NUt<f4GcB+-JpYWz+ zs8;uIoxuUh&3Rsuj#Uy~;(qP&f-zFcto%BTbdLPhmSKVHZM`YI5PP;{wn%vS+TxAb z`n-i9T>dDf0YnRN*=pffVwvv}Gg)-iO?K8n{vk*a0}D_d_VPQUUX$UXt)}wk8&WF4 zm;aQ5eJl7UZ0w9aAS;ZCciD!p-p_RG{^=XVg=KQD)=q=gh`rO5I0$~N;K%aIb*VVu zWE9H2nOdbBCJGEt;h+>qpL|(|G&R>`197z^VSHybGJKJ`i?x_^q_p1Pqm&b@xM;Q zLgeQ%soKi2-%B>4-md;aNt8wind&--b(l;-REG)N;P-odKm9;|bk!Uo<9Ig0uI$e{ za3ww~OuBJ~7I3FHg<8g}Da+ON8GOnB&kgUQ!qLvR*xGIe0x(`MRbjSy?HCsa5I&^U zEkLPwp@I(n^R)QnBT`Y72gpOm6aVF2&uEQ{ar7n2{P_8S+ay7`{FA*jUulm&mNf}{ z*Pz`H19mNZ!52BH8wz~tt>20!s986QSL>$lskP#Tb58$J>e3jk3#BS0F@kh{mDlCB zE$rbv(A0`}%NzfLBgQykn&mDrlelZY*LjU{23yC~~3DRmm_{ey%IhJ=kN zq|8TQ5TDA+#RRMPc&17M-rw0cZSJdy-(57f>OT1nZa8r={-)!M>pS}(V`S?e6EDZ& zQd{I>{*8(@Xei&{ zg|g`zZY3ad#><#HWL6T|r}o*Gx$=j-Gf4*>y|6E}-IDt3AN4lqNR}X=0m7qe6TVa{ zPVMS#E%7(Z*aww);?bY$v|a#HY>Rqua&|njv57l$|Mbo{fS2kxxRd~Vz?}jJU;%W0 z=kG5C0swJ$?KSzoBnE^hFFAbdU9bRg_yOXaRA--8a~7U-{<5I)Vh>0@a*rY;D#zaU zMrmr5REqv4$&Vcg-?zD#_FkQeYiBTFnReg{eShZa@VhDgz$LuJmK*)5NpG5D^wIJB zhoyT(hJRL$@|2kR5fXZLypN;&DH*Y47+=q4@Xb}=hKw3ahvdA^A~S2{od~V z@J=(UWU7@VT=&w8u8v1Zr{wbGC+C7IAJ_IzPceP!`KGL(#`d^%5L zn26X;*1s*!qS#`oq5_Fg+T32uz546&*=cg3r}<4fUsrnm<@kIJBJ95s5rwvIbT9!W zviic}UeEv~;8}efws?vPxZs8LC^G=j(;m(Nu8lH0m2c3mcwf!{M1oIT^Ip@%zwB0E z6^Uk}a4?F%jQrZ1}SI}NBle>d?o|~L=J@DK$t-u)BlU1ca`D=0s!v+pDniBhztlT V-TyD8|HNPrzaXsmGkpKd{1=R7A({XH diff --git a/doc/sphinx/03_vibe/cpu_20.csv b/doc/sphinx/03_vibe/cpu_20.csv index 459ecb2d..5ba4e243 100644 --- a/doc/sphinx/03_vibe/cpu_20.csv +++ b/doc/sphinx/03_vibe/cpu_20.csv @@ -1,4 +1,4 @@ No. Cores, Actual, Ideal -8, 2.00e+06, 2.0e+06 -32, 7.40e+06, 8.0e+06 -56, 1.29e+07, 1.4e+07 \ No newline at end of file +8, 3.40e+06, 3.40e+06 +32, 1.19e+07, 1.36e+07 +56, 1.88e+07, 2.38e+07 \ No newline at end of file diff --git a/doc/sphinx/03_vibe/cpu_40.csv b/doc/sphinx/03_vibe/cpu_40.csv index ccc095c0..96f511dd 100644 --- a/doc/sphinx/03_vibe/cpu_40.csv +++ b/doc/sphinx/03_vibe/cpu_40.csv @@ -1,6 +1,6 @@ No. Cores, Actual, Ideal -8, 1.82e+06, 1.82e+06 -32, 7.04e+06, 7.28e+06 -56, 1.21e+07, 1.274e+07 -88, 1.60e+07, 2.02e+07 -112, 2.00e+07, 2.548e+07 +8, 2.80e+06, 2.80e+06 +32, 1.12e+07, 1.12e+07 +56, 1.79e+07, 1.96e+07 +88, 2.36e+07, 3.08e+07 +112, 2.61e+07, 3.92e+07 diff --git a/doc/sphinx/03_vibe/cpu_60.csv b/doc/sphinx/03_vibe/cpu_60.csv index 5cb43156..9f7e8829 100644 --- a/doc/sphinx/03_vibe/cpu_60.csv +++ b/doc/sphinx/03_vibe/cpu_60.csv @@ -1,6 +1,6 @@ No. Cores, Actual, Ideal -8, 1.51e+06, 1.51e+06 -32, 6.34e+06, 6.04e+06 -56, 1.09e+07, 1.057e+07 -88, 1.55e+07, 1.661e+07 -112, 1.85e+07, 2.114e+07 +8, 2.40e+06, 2.40e+06 +32, 9.56e+06, 9.60e+06 +56, 1.54e+07, 1.68e+07 +88, 2.16e+07, 2.64e+07 +112, 2.44e+07, 3.36e+07 \ No newline at end of file diff --git a/doc/sphinx/03_vibe/gpu.csv b/doc/sphinx/03_vibe/gpu.csv index 96abcfbd..69c8b470 100644 --- a/doc/sphinx/03_vibe/gpu.csv +++ b/doc/sphinx/03_vibe/gpu.csv @@ -1,7 +1,7 @@ Mesh Base Size, Actual -32, 1.75e+07 -64, 1.15e+07 -96, 6.78e+06 -128, 0 -160, 0 +32, 2.88e+07 +64, 2.19e+07 +96, 1.41e+07 +128, 1.36e+07 +160, 1.03e+07 192, 0 diff --git a/doc/sphinx/03_vibe/vibe.rst b/doc/sphinx/03_vibe/vibe.rst index 862eaf2d..da5c8f1d 100644 --- a/doc/sphinx/03_vibe/vibe.rst +++ b/doc/sphinx/03_vibe/vibe.rst @@ -67,8 +67,7 @@ To build Parthenon on CPU, including this benchmark, with minimal external depen .. code-block:: bash parthenon$ mkdir build && cd build - build$ export CXXFLAGS="-fno-math-errno -march=native" - build$ cmake -DPARTHENON_DISABLE_HDF5=ON -DPARTHENON_ENABLE_PYTHON_MODULE_CHECK=OFF -DREGRESSION_GOLD_STANDARD_SYNC=OFF -DCMAKE_BUILD_TYPE=Release ../ + build$ cmake -DPARTHENON_DISABLE_HDF5=ON -DPARTHENON_ENABLE_PYTHON_MODULE_CHECK=OFF -DREGRESSION_GOLD_STANDARD_SYNC=OFF -DPARTHENON_ENABLE_TESTING=OFF -DCMAKE_BUILD_TYPE=Release ../ build$ make -j .. @@ -81,11 +80,11 @@ On Crossroads the relevant modules for the results shown here are: .. -To build for execution on a single GPU, it should be sufficient to add the following flags to the CMake configuration line +To build for execution on a single GPU, it should be sufficient to add flags similar to the CMake configuration line .. code-block:: bash - cmake -DPARTHENON_DISABLE_MPI=ON -DKokkos_ENABLE_CUDA=ON -DKokkos_ARCH_AMPERE80=ON + cmake -DKokkos_ENABLE_CUDA=ON -DKokkos_ARCH_AMPERE80=ON .. @@ -123,7 +122,7 @@ The results presented here use 128 and 160 for memory footprints of approximate Results from Parthenon are provided on the following systems: * Crossroads (see :ref:`GlobalSystemATS3`) -* An Nvidia A100 GPU hosted on an [Nvidia Arm HPC Developer Kit](https://developer.nvidia.com/arm-hpc-devkit) +* A Grace Hopper (Grace ARM CPU 72 cores with 120GB, H100 GPU with 96GB) The mesh and meshblock size parameters are chosen to balance realism/performance with memory footprint. For the following tests we @@ -182,12 +181,12 @@ Crossroads VIBE Throughput Performance on Crossroads using ~60% memory -Nvidia testbed with A100 +Nvidia Grace Hopper ------------------------ -Throughput performance of Parthenon-VIBE on a 40GB A100 is provided within the following table and figure. +Throughput performance of Parthenon-VIBE on a 96 GB H100 is provided within the following table and figure. -.. csv-table:: VIBE Throughput Performance on A100 +.. csv-table:: VIBE Throughput Performance on H100 :file: gpu.csv :align: center :widths: 10, 10 @@ -196,9 +195,9 @@ Throughput performance of Parthenon-VIBE on a 40GB A100 is provided within the f .. figure:: gpu.png :align: center :scale: 50% - :alt: VIBE Throughput Performance on A100 + :alt: VIBE Throughput Performance on H100 - VIBE Throughput Performance on A100 + VIBE Throughput Performance on H100 Multi-node scaling on Crossroads