High Performance Computing of Complex Electromagnetic Algorithms Based on GPU/CPU Heterogeneous Platform and Its Applications to EM Scattering and Multilayered Medium Structure

<table class="table-group" id="tab3"><tr><td><table class="table"><tr><td class="thead-hr" colspan="2"><hr/></td></tr><tr class="thead"><td class="align_left">Variables</td><td class="align_left">Description</td></tr><tr><td class="thead-hr" colspan="2"><hr/></td></tr><tr><td class="align_left"><span style="width: 14.0706ptpx;"><svg height="10.9953pt" id="M75" style="vertical-align:-3.18148pt" version="1.1" viewbox="-0.0498162 -7.81382 14.0706 10.9953" width="14.0706pt" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink"><g transform="matrix(.013,0,0,-0.013,0,0)"><path d="M324 430H196L233 583L223 592L145 529L120 430H54L29 396L31 388H111L56 126C33 15 54 -12 77 -12C137 -12 214 57 250 95L233 119C208 92 155 59 138 59C126 59 120 70 131 125L186 390L298 394L324 430Z" id="g113-117"></path><glyph.data ascent="3473" descent="-2876" horiz-adv-x="341" vert-adv-y="341" xpluscharwidth="4.433"></glyph.data></g><g transform="matrix(.0091,0,0,-0.0091,4.368,3.132)"><path d="M1006 650H829L492 147L357 650H174L166 618C270 609 275 601 247 502L194 315C158 189 136 127 122 96C104 56 80 37 18 32L12 0H256L263 32C181 40 176 51 183 98C188 132 197 179 226 281L299 541H301L433 6H458L809 551H811L753 141C739 41 734 39 650 32L642 0H917L923 32C836 39 827 41 839 142L887 535C896 610 901 612 997 618L1006 650Z" id="g50-78"></path><glyph.data ascent="3443" descent="-2856" horiz-adv-x="1030" vert-adv-y="1030" xpluscharwidth="13.740969100000001"></glyph.data></g></svg></span></td><td class="align_left">Time for near matrix generation</td></tr><tr><td class="align_left"><span style="width: 9.34824ptpx;"><svg height="11.1049pt" id="M76" style="vertical-align:-3.29108pt" version="1.1" viewbox="-0.0498162 -7.81382 9.34824 11.1049" width="9.34824pt" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink"><g transform="matrix(.013,0,0,-0.013,0,0)"><path d="M324 430H196L233 583L223 592L145 529L120 430H54L29 396L31 388H111L56 126C33 15 54 -12 77 -12C137 -12 214 57 250 95L233 119C208 92 155 59 138 59C126 59 120 70 131 125L186 390L298 394L324 430Z" id="g113-117"></path><glyph.data ascent="3473" descent="-2876" horiz-adv-x="341" vert-adv-y="341" xpluscharwidth="4.433"></glyph.data></g><g transform="matrix(.0091,0,0,-0.0091,4.368,3.132)"><path d="M463 633C453 636 427 644 417 648C388 660 349 667 316 667C183 667 100 592 100 483C100 398 176 335 223 305L259 282C321 242 358 203 358 149C358 66 301 22 234 22C110 22 70 131 57 208L24 205L28 48C51 23 128 -17 215 -17C342 -17 446 60 446 174C446 262 394 310 317 361L284 383C229 420 183 455 183 515C183 569 222 628 298 628C384 628 418 563 428 481L461 487C459 527 458 593 463 633Z" id="g50-84"></path><glyph.data ascent="3443" descent="-2856" horiz-adv-x="487" vert-adv-y="487" xpluscharwidth="8.79968539"></glyph.data></g></svg></span></td><td class="align_left">Time for corresponding SpMVP</td></tr><tr><td class="align_left"><span style="width: 16.3358ptpx;"><svg height="10.9953pt" id="M77" style="vertical-align:-3.18148pt" version="1.1" viewbox="-0.0498162 -7.81382 16.3358 10.9953" width="16.3358pt" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink"><g transform="matrix(.013,0,0,-0.013,0,0)"><path d="M324 430H196L233 583L223 592L145 529L120 430H54L29 396L31 388H111L56 126C33 15 54 -12 77 -12C137 -12 214 57 250 95L233 119C208 92 155 59 138 59C126 59 120 70 131 125L186 390L298 394L324 430Z" id="g113-117"></path><glyph.data ascent="3473" descent="-2876" horiz-adv-x="341" vert-adv-y="341" xpluscharwidth="4.433"></glyph.data></g><g transform="matrix(.0091,0,0,-0.0091,4.368,3.132)"><path d="M682 0V33C615 38 603 49 571 134C507 305 441 494 378 665L343 655L136 136C101 48 86 41 22 33V0H246V33C170 41 161 50 183 112C196 152 209 192 224 232H441C461 176 478 126 493 87C504 49 499 41 433 33V0H682ZM425 279H242C273 363 303 448 335 533H337L425 279Z" id="g58-63"></path><glyph.data ascent="1007" descent="-369" horiz-adv-x="703" vert-adv-y="703" xpluscharwidth="10.765278910000001"></glyph.data></g><g transform="matrix(.0091,0,0,-0.0091,10.765,3.132)"><path d="M502 500C497 551 492 613 491 650H43V617C122 612 130 606 130 521V129C130 45 122 38 41 33V0H319V33C227 38 217 45 217 129V305H319C411 305 413 295 427 225H461V421H427C414 357 411 348 319 348H217V580C217 609 219 612 250 612H330C401 612 423 605 439 578C452 557 458 539 467 496L502 500Z" id="g58-68"></path><glyph.data ascent="1007" descent="-369" horiz-adv-x="540" vert-adv-y="540" xpluscharwidth="15.678983800000001"></glyph.data></g></svg></span></td><td class="align_left">Time for aggregation at finest level</td></tr><tr><td class="align_left"><span style="width: 16.6464ptpx;"><svg height="10.9953pt" id="M78" style="vertical-align:-3.18148pt" version="1.1" viewbox="-0.0498162 -7.81382 16.6464 10.9953" width="16.6464pt" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink"><g transform="matrix(.013,0,0,-0.013,0,0)"><path d="M324 430H196L233 583L223 592L145 529L120 430H54L29 396L31 388H111L56 126C33 15 54 -12 77 -12C137 -12 214 57 250 95L233 119C208 92 155 59 138 59C126 59 120 70 131 125L186 390L298 394L324 430Z" id="g113-117"></path><glyph.data ascent="3473" descent="-2876" horiz-adv-x="341" vert-adv-y="341" xpluscharwidth="4.433"></glyph.data></g><g transform="matrix(.0091,0,0,-0.0091,4.368,3.132)"><path d="M43 650V617C122 612 130 606 130 522V128C130 44 122 37 34 32V0H275C400 0 497 25 573 82C652 142 698 229 698 350C698 445 662 518 613 565C549 625 456 650 327 650H43ZM217 544C217 580 220 591 229 599C240 607 264 613 311 613C380 613 437 599 482 570C561 523 598 436 598 336C598 177 511 40 324 40C248 40 217 61 217 135V544Z" id="g58-66"></path><glyph.data ascent="1007" descent="-369" horiz-adv-x="744" vert-adv-y="744" xpluscharwidth="11.138377680000001"></glyph.data></g><g transform="matrix(.0091,0,0,-0.0091,11.074,3.132)"><path d="M502 500C497 551 492 613 491 650H43V617C122 612 130 606 130 521V129C130 45 122 38 41 33V0H319V33C227 38 217 45 217 129V305H319C411 305 413 295 427 225H461V421H427C414 357 411 348 319 348H217V580C217 609 219 612 250 612H330C401 612 423 605 439 578C452 557 458 539 467 496L502 500Z" id="g58-68"></path><glyph.data ascent="1007" descent="-369" horiz-adv-x="540" vert-adv-y="540" xpluscharwidth="15.987983799999999"></glyph.data></g></svg></span></td><td class="align_left">Time for disaggregation at finest level</td></tr><tr><td class="align_left"><span style="width: 15.7786ptpx;"><svg height="11.1095pt" id="M79" style="vertical-align:-3.29568pt" version="1.1" viewbox="-0.0498162 -7.81382 15.7786 11.1095" width="15.7786pt" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink"><g transform="matrix(.013,0,0,-0.013,0,0)"><path d="M324 430H196L233 583L223 592L145 529L120 430H54L29 396L31 388H111L56 126C33 15 54 -12 77 -12C137 -12 214 57 250 95L233 119C208 92 155 59 138 59C126 59 120 70 131 125L186 390L298 394L324 430Z" id="g113-117"></path><glyph.data ascent="3473" descent="-2876" horiz-adv-x="341" vert-adv-y="341" xpluscharwidth="4.433"></glyph.data></g><g transform="matrix(.0091,0,0,-0.0091,4.368,3.132)"><path d="M682 0V33C615 38 603 49 571 134C507 305 441 494 378 665L343 655L136 136C101 48 86 41 22 33V0H246V33C170 41 161 50 183 112C196 152 209 192 224 232H441C461 176 478 126 493 87C504 49 499 41 433 33V0H682ZM425 279H242C273 363 303 448 335 533H337L425 279Z" id="g58-63"></path><glyph.data ascent="1007" descent="-369" horiz-adv-x="703" vert-adv-y="703" xpluscharwidth="10.765278910000001"></glyph.data></g><g transform="matrix(.0091,0,0,-0.0091,10.72,3.132)"><path d="M416 501L399 642C356 655 314 665 271 665C140 665 63 582 63 486C63 372 163 325 229 290C304 250 359 217 359 142C359 70 316 25 240 25C138 25 94 125 78 186L44 178C52 123 64 45 72 21C81 16 96 8 121 0S179 -15 220 -15C353 -15 445 68 445 176C445 289 350 335 270 375C191 414 143 449 143 520C143 572 177 627 253 627C340 627 366 556 382 495L416 501Z" id="g58-81"></path><glyph.data ascent="1007" descent="-369" horiz-adv-x="484" vert-adv-y="484" xpluscharwidth="15.12438548"></glyph.data></g></svg></span></td><td class="align_left">Time for aggregation towards coarsest level</td></tr><tr><td class="align_left"><span style="width: 16.1988ptpx;"><svg height="11.1095pt" id="M80" style="vertical-align:-3.29568pt" version="1.1" viewbox="-0.0498162 -7.81382 16.1988 11.1095" width="16.1988pt" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink"><g transform="matrix(.013,0,0,-0.013,0,0)"><path d="M324 430H196L233 583L223 592L145 529L120 430H54L29 396L31 388H111L56 126C33 15 54 -12 77 -12C137 -12 214 57 250 95L233 119C208 92 155 59 138 59C126 59 120 70 131 125L186 390L298 394L324 430Z" id="g113-117"></path><glyph.data ascent="3473" descent="-2876" horiz-adv-x="341" vert-adv-y="341" xpluscharwidth="4.433"></glyph.data></g><g transform="matrix(.0091,0,0,-0.0091,4.368,3.132)"><path d="M43 650V617C122 612 130 606 130 522V128C130 44 122 37 34 32V0H275C400 0 497 25 573 82C652 142 698 229 698 350C698 445 662 518 613 565C549 625 456 650 327 650H43ZM217 544C217 580 220 591 229 599C240 607 264 613 311 613C380 613 437 599 482 570C561 523 598 436 598 336C598 177 511 40 324 40C248 40 217 61 217 135V544Z" id="g58-66"></path><glyph.data ascent="1007" descent="-369" horiz-adv-x="744" vert-adv-y="744" xpluscharwidth="11.138377680000001"></glyph.data></g><g transform="matrix(.0091,0,0,-0.0091,11.138,3.132)"><path d="M416 501L399 642C356 655 314 665 271 665C140 665 63 582 63 486C63 372 163 325 229 290C304 250 359 217 359 142C359 70 316 25 240 25C138 25 94 125 78 186L44 178C52 123 64 45 72 21C81 16 96 8 121 0S179 -15 220 -15C353 -15 445 68 445 176C445 289 350 335 270 375C191 414 143 449 143 520C143 572 177 627 253 627C340 627 366 556 382 495L416 501Z" id="g58-81"></path><glyph.data ascent="1007" descent="-369" horiz-adv-x="484" vert-adv-y="484" xpluscharwidth="15.54238548"></glyph.data></g></svg></span></td><td class="align_left">Time for disaggregation towards finest level</td></tr><tr class="table-tr"><td colspan="2"><hr class="tbody-hr"/></td></tr></table></td></tr></table>

Variables definition of time consumption.

International Journal of Antennas and Propagation

tab3

Table 3

Table 3: High Performance Computing of Complex Electromagnetic Algorithms Based on GPU/CPU Heterogeneous Platform and Its Applications to EM Scattering and Multilayered Medium Structure 

Table 3 | High Performance Computing of Complex Electromagnetic Algorithms Based on GPU/CPU Heterogeneous Platform and Its Applications to EM Scattering and Multilayered Medium Structure 