@@ -253,15 +253,15 @@ <h1><a class="anchor" id="autotoc_md81"></a>
253253< h2 > < a class ="anchor " id ="autotoc_md82 "> </ a >
254254AMD MI250X GPU</ h2 >
255255< p > MFC weask scales to (at least) 65,536 AMD MI250X GPUs on OLCF Frontier with 96% efficiency. This corresponds to 87% of the entire machine.</ p >
256- < p > < img src ="../res/weakScaling/frontier.svg " alt ="" style ="pointer-events: none; height: 50%; width:50%; border-radius: 10ptclass ="inline "/> </ p >
256+ < p > < img src ="../res/weakScaling/frontier.svg " alt ="" style ="height: 50%; width:50%; border-radius: 10pt pointer-events: none; " class ="inline "/> </ p >
257257< h2 > < a class ="anchor " id ="autotoc_md83 "> </ a >
258258NVIDIA V100 GPU</ h2 >
259259< p > MFC weak scales to (at least) 13,824 V100 NVIDIA V100 GPUs on OLCF Summit with 97% efficiency. This corresponds to 50% of the entire machine.</ p >
260- < p > < img src ="../res/weakScaling/summit.svg " alt ="" style ="pointer-events: none; height: 50%; width:50%; border-radius: 10ptclass ="inline "/> </ p >
260+ < p > < img src ="../res/weakScaling/summit.svg " alt ="" style ="height: 50%; width:50%; border-radius: 10pt pointer-events: none; " class ="inline "/> </ p >
261261< h2 > < a class ="anchor " id ="autotoc_md84 "> </ a >
262262IBM Power9 CPU</ h2 >
263263< p > MFC Weak scales to 13,824 Power9 CPU cores on OLCF Summit to within 1% of ideal scaling.</ p >
264- < p > < img src ="../res/weakScaling/cpuScaling.svg " alt ="" style ="pointer-events: none; height: 50%; width:50%; border-radius: 10ptclass ="inline "/> </ p >
264+ < p > < img src ="../res/weakScaling/cpuScaling.svg " alt ="" style ="height: 50%; width:50%; border-radius: 10pt pointer-events: none; " class ="inline "/> </ p >
265265< h1 > < a class ="anchor " id ="autotoc_md85 "> </ a >
266266Strong scaling</ h1 >
267267< p > Strong scaling results are obtained by keeping the problem size constant and increasing the number of processes so that work per process decreases.</ p >
@@ -270,14 +270,14 @@ <h2><a class="anchor" id="autotoc_md86"></a>
270270< p > The base case utilizes 8 GPUs with one MPI process per GPU for these tests. The performance is analyzed at two problem sizes: 16M and 64M grid points. The "base case" uses 2M and 8M grid points per process.</ p >
271271< h3 > < a class ="anchor " id ="autotoc_md87 "> </ a >
27227216M Grid Points</ h3 >
273- < p > < img src ="../res/strongScaling/strongScaling16.svg " alt ="" style ="pointer-events: none; width: 50%; border-radius: 10ptclass ="inline "/> </ p >
273+ < p > < img src ="../res/strongScaling/strongScaling16.svg " alt ="" style ="width: 50%; border-radius: 10pt pointer-events: none; " class ="inline "/> </ p >
274274< h3 > < a class ="anchor " id ="autotoc_md88 "> </ a >
27527564M Grid Points</ h3 >
276- < p > < img src ="../res/strongScaling/strongScaling64.svg " alt ="" style ="pointer-events: none; width: 50%; border-radius: 10ptclass ="inline "/> </ p >
276+ < p > < img src ="../res/strongScaling/strongScaling64.svg " alt ="" style ="width: 50%; border-radius: 10pt pointer-events: none; " class ="inline "/> </ p >
277277< h2 > < a class ="anchor " id ="autotoc_md89 "> </ a >
278278IBM Power9 CPU</ h2 >
279279< p > CPU strong scaling tests are done with problem sizes of 16, 32, and 64M grid points, with the base case using 2, 4, and 8M cells per process.</ p >
280- < p > < img src ="../res/strongScaling/cpuStrongScaling.svg " alt ="" style ="pointer-events: none; width: 50%; border-radius: 10ptclass ="inline "/> </ p >
280+ < p > < img src ="../res/strongScaling/cpuStrongScaling.svg " alt ="" style ="width: 50%; border-radius: 10pt pointer-events: none; " class ="inline "/> </ p >
281281</ div > </ div > <!-- contents -->
282282</ div > <!-- PageDoc -->
283283</ div > <!-- doc-content -->
0 commit comments