[ MeP \u306e\u62e1\u5f35\u6a5f\u80fd\u3092\u7528\u3044\u305f\u518d\u69cb\u6210\u578b\u30d7\u30ed\u30bb\u30c3\u30b5Hy-DiSC\u306e\u6027\u80fd\u8a55\u4fa1 ]
\nMeP \u3067\u9010\u6b21\u51e6\u7406\u3001DS-HIE \u30d7\u30ed\u30bb\u30c3\u30b5\u3067\u30b9\u30c8\u30ea\u30fc\u30df\u30f3\u30b0\u51e6\u7406\u3002
\n\u30c7\u30fc\u30bf\u4f9b\u7d66\u306a\u3069\u306f MeP \u304c\u884c\u3046\u306e\u3067\u3001\u30ad\u30e3\u30c3\u30b7\u30e5\u30df\u30b9\u304c\u6027\u80fd\u4f4e\u4e0b\u306e\u8981\u56e0\u306b\u306a\u308b\u306e\u3067\u3001\u30e1\u30e2\u30ea\u30a2\u30af\u30bb\u30b9\u3092\u542b\u3093\u3060\u90e8\u5206\u306e\u30b7\u30df\u30e5\u30ec\u30fc\u30bf\u3092\u69cb\u7bc9\u3002
\n\u8a55\u4fa1\u30a2\u30d7\u30ea\u30b1\u30fc\u30b7\u30e7\u30f3\u306f JPEG encoder \u306e DCT \u3068\u304b\u3002
\n\u30ad\u30e3\u30c3\u30b7\u30e5\u306f 2-way set associative.
\n[ \u52d5\u7684\u30ea\u30b3\u30f3\u30d5\u30a3\u30ae\u30e3\u30e9\u30d6\u30eb\u30d7\u30ed\u30bb\u30c3\u30b5 MuCCRA-3 \u306e\u5b9f\u6a5f\u8a55\u4fa1 ]
\nDR \u306a\u96fb\u529b\u8a55\u4fa1\u306f\u306a\u304b\u306a\u304b\u8a73\u7d30\u304c\u51fa\u3066\u3053\u306a\u3044\u306e\u3067\u3001\u3084\u3063\u3066\u307f\u308b\u3002
\nInterconnect \u306f\u3001island style \u3068\u76f4\u7d50\u306e\u30cf\u30a4\u30d6\u30ea\u30c3\u30c9\u3002
\nSMU \u3084\u3081\u307e\u3057\u305f\u3002ALU \u3068\u30ec\u30b8\u30b9\u30bf\u30d5\u30a1\u30a4\u30eb\u3060\u3051\u3002SMU \u306e\u304b\u308f\u308a\u306b\u3001\u30b7\u30d5\u30c8\u3068\u304b\u3092\u3084\u308b\u305f\u3081\u306e\u56de\u8def\u304cALU\u306e\u5165\u308a\u53e3\u306b\u3064\u3044\u3066\u3044\u308b\u3002
\n\u03b1 blender @ 41MHz \u3067 11mW.
\n\u30c7\u30fc\u30bf\u30d1\u30b9\u306e\u5909\u66f4\u3068\u304b\u30af\u30ed\u30c3\u30af\u30c4\u30ea\u30fc\u306e\u96fb\u529b\u304c\u5927\u304d\u3044\u3002
\n\u30b3\u30f3\u30c6\u30ad\u30b9\u30c8\u6570\u306f? (\u4e95\u53e3\u5148\u751f) \u2192 32\u30b3\u30f3\u30c6\u30ad\u30b9\u30c8\u6301\u3063\u3066\u3044\u308b\u3002
\n4×4 \u3067\u8a55\u4fa1\u3057\u3066\u3044\u308b\u304c\u3001\u3082\u3063\u3068\u5927\u304d\u3044\u306e\u3067\u3084\u3063\u305f\u5834\u5408\u306e\u96fb\u529b\u8a55\u4fa1\u306e\u7d50\u679c\u306f\u3069\u3046\u306a\u308b\u3068\u8003\u3048\u3089\u308c\u308b? (\u5f18\u4e2d\u5148\u751f) \u2192 \u5f85\u6a5f\u30fb\u5b9f\u884c\u6642\u306e\u96fb\u529b\u3092\u5206\u985e\u3057\u305f\u306e\u3067\u3001PE \u6570\u304c\u5897\u3048\u3066\u3082\u5272\u5408\u306f\u5909\u308f\u3089\u306a\u3044\u3068\u601d\u3046\u3002\u30ea\u30fc\u30af\u306f 0.6mA \u3067\u3059\u3054\u304f\u5c0f\u3055\u3044\u306e\u3067\u3001\u9762\u7a4d\u304c\u5927\u304d\u304f\u306a\u3063\u3066\u3082\u30ea\u30fc\u30af\u304c\u5927\u5909\u3001\u3068\u3044\u3046\u3053\u3068\u306f\u306a\u3055\u305d\u3046\u3002
\n\u3055\u307e\u3056\u307e\u306a\u30a2\u30fc\u30ad\u30c6\u30af\u30c1\u30e3\u306e\u5909\u66f4\u3068\u6d88\u8cbb\u96fb\u529b\u524a\u6e1b\u306e\u95a2\u4fc2\u306f (\u30d1\u30ca\u30bd\u30cb\u30c3\u30af\u4e0a\u7530\u3055\u3093) \u2192 \u3044\u307e\u307e\u3067\u3042\u307e\u308a\u5229\u7528\u3055\u308c\u3066\u3044\u306a\u304b\u3063\u305f\u3053\u3068\u3092\u524a\u3063\u305f\u3068\u3053\u308d\u304c\u52b9\u3044\u3066\u3044\u308b\u3068\u601d\u3046\u3002\u30d7\u30ed\u30bb\u30b9\u304c\u9055\u3046\u306e\u3067\u5358\u7d14\u306a\u6bd4\u8f03\u304c\u3067\u304d\u306a\u3044\u3067\u3059\u3001\u3059\u3044\u307e\u305b\u3093\u3002
\n\u6027\u80fd\u306f? \u2192 \u4eca\u56de\u306f\u8a55\u4fa1\u3092\u3068\u3063\u3066\u3044\u306a\u3044\u306e\u3067\u3001\u4f55\u3068\u3082\u3002\u30ec\u30b8\u30b9\u30bf\u3092\u5165\u308c\u305f\u306e\u3067\u30af\u30ed\u30c3\u30af\u6570\u306f\u5897\u3048\u308b\u50be\u5411\u3060\u3051\u3069\u300110% \u304b 20% \u304f\u3089\u3044\u3002
\n[ Performance and Cost Evaluations of On-Chip Network Topologies in FPGAs ]
\n2D-Mesh, 2D-Torus, Spidergon, Fat-Tree, Concentrated Mesh \u306e5\u7a2e\u985e\u306e\u30c8\u30dd\u30ed\u30b8\u3092\u8a55\u4fa1\u3002
\nSpidergon \u306f bidirectional ring \u3060\u3051\u3069\u3001\u5404\u30ce\u30fc\u30c9\u304c\u53cd\u5bfe\u5074\u306e\u30ce\u30fc\u30c9\u3078\u30b7\u30e7\u30fc\u30c8\u30ab\u30c3\u30c8\u3067\u304d\u308b\u7d4c\u8def\u3092\u3082\u3064\u3002
\nHardware amount (area), bandwidth, latency \u3092\u8a55\u4fa1\u3002
\n\u5bfe\u8c61\u30c7\u30d0\u30a4\u30b9\u306f V4LX200.
\n\u30c8\u30e9\u30d5\u30a3\u30c3\u30af\u30d1\u30bf\u30fc\u30f3\u306f\u3001\u306a\u3059\u3071\u3089\u3001\u3067\u3044\u3044\u306e? (\u67f4\u7530\u5148\u751f) \u2192 Yes, HPC on FPGA \u3092\u60f3\u5b9a\u3057\u3066\u3044\u307e\u3059\u3002
\n[ \u30b9\u30e2\u30fc\u30eb\u30ef\u30fc\u30eb\u30c9\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u5316\u914d\u7dda\u69cb\u9020\u306b\u3088\u308b FPGA \u306e\u6d88\u8cbb\u96fb\u529b\u524a\u6e1b ]
\nFPGA\u306f\u914d\u7dda\u30ea\u30bd\u30fc\u30b9\u304c\u305f\u304f\u3055\u3093\u5fc5\u8981\u3002
\nSWN \u306b\u3057\u3066\u3001\u4f4e\u8ca0\u8377\u5bb9\u91cf\u306e\u914d\u7dda\u7d4c\u8def\u3092\u4f5c\u308b\u3002Activity \u3068\u8ca0\u8377\u5bb9\u91cf\u3092\u8003\u616e\u3057\u305f\u914d\u7dda\u3002
\nActivity \u304c\u9ad8\u3044 net \u306f\u8ca0\u8377\u5bb9\u91cf\u306e\u4f4e\u3044\u914d\u7dda\u7d4c\u8def\u306b\u8f09\u305b\u305f\u3044\u3002SW \u30e9\u30a4\u30f3\u306f\u3001\u901a\u904e\u3059\u308b\u30b9\u30a4\u30c3\u30c1\u30d6\u30ed\u30c3\u30af\u304c\u5c11\u306a\u304f\u3001\u659c\u3081\u914d\u7dda\u3067\u914d\u7dda\u3082\u77ed\u3044\u306e\u3067\u3001\u8ca0\u8377\u5bb9\u91cf\u304c\u5c0f\u3055\u304f\u306a\u308b\u3002
\nVPR \u306e\u30b3\u30b9\u30c8\u95a2\u6570\u306b activity x capacity \u306e\u9805\u3092\u8ffd\u52a0\u3057\u305f\u3002P-VPR (power aware VPR) \u3092\u5143\u306b\u3057\u3066\u3044\u308b\u3002\u30d9\u30f3\u30c1\u30de\u30fc\u30af\u306f MCNC. Regular \u306b\u6bd4\u3079\u3066\u5e73\u5747 8.3% \u304f\u3089\u3044\u96fb\u529b\u3092\u6e1b\u3089\u305b\u308b (static power \u304c\u307b\u3093\u306e\u3059\u3053\u3057\u5897\u3048\u3066\u3001dynamic power \u304c\u6e1b\u308b)\u3002\u9045\u5ef6\u306f\u6e1b\u308b\u3053\u3068\u3082\u3042\u308b\u3051\u3069\u3001\u57fa\u672c\u7684\u306b\u4f38\u3073\u3066\u3057\u307e\u3046 (\u306a\u3093\u3067?? long line \u524a\u3063\u305f\u304b\u3089???)\u3002
\nstatic power \u304c dynamic \u306b\u6bd4\u3079\u3066\u3072\u3068\u3051\u305f\u304f\u3089\u3044\u5c0f\u3055\u3044\u306e\u306f\u306a\u305c\uff1f V4 \u3068\u304b\u3060\u3068\u300140% \u304f\u3089\u3044\u304c static \u3067\u3059\u304c\u3002\u30d7\u30ed\u30bb\u30b9\u306e\u305b\u3044\u306a\u306e\u304b\u3057\u3089 (\u5929\u91ce\u5148\u751f)? \u2192 \u8a55\u4fa1\u3068\u308b\u3068\u304d\u306b\u3001\u3042\u308c? \u3068\u3044\u3046\u30dd\u30a4\u30f3\u30c8\u304c\u3042\u3063\u305f\u306e\u3067\u3001\u3082\u3057\u304b\u3057\u305f\u3089\u6b63\u3057\u304f\u306a\u3044\u304b\u3082…
\nDynamic \u306f\u659c\u3081\u7dda\u3067\u77ed\u304f\u3057\u305f\u306e\u304c\u304d\u3044\u3066\u3044\u308b\u306e\u304b\u3001\u30e9\u30f3\u30c0\u30e0\u306b\u30b7\u30e7\u30fc\u30c8\u30ab\u30c3\u30c8\u3067\u304d\u308b\u3053\u3068\u304c\u52b9\u3044\u3066\u3044\u308b\u306e\u304b (\u5929\u91ce\u5148\u751f)? \u2192 \u901a\u904e\u3059\u308b\u30b9\u30a4\u30c3\u30c1\u304c\u6e1b\u3089\u305b\u308b\u70b9\u304c\u5927\u304d\u3044\u306e\u3060\u3068\u601d\u3044\u307e\u3059\u3002
\n\u659c\u3081\u914d\u7dda\u306f\u30e9\u30f3\u30c0\u30e0\u306b\u3064\u3051\u308b\u3088\u3046\u3060\u304c\u3001\u898f\u5247\u6b63\u3057\u304f\u659c\u3081\u306b\u5165\u308c\u308b\u306e\u3068\u6bd4\u3079\u308b\u3068\u3069\u3046? \u914d\u7dda\u30ec\u30a4\u30e4\u304c\u5897\u3048\u308b\u3053\u3068\u306b\u306f\u5909\u308f\u308a\u306a\u3044\u3088\u306d (\u4e95\u53e3\u5148\u751f@JAIST)? \u2192 \u4ee5\u524d\u3084\u3063\u305f\u8a55\u4fa1\u3067\u306f\u3069\u3061\u3089\u3067\u3082\u3042\u307e\u308a\u304b\u308f\u308a\u306a\u304b\u3063\u305f\u3067\u3059…
\n\u9045\u5ef6\u304c\u3061\u3087\u3063\u3068\u5927\u304d\u304f\u306a\u3063\u3066\u3044\u308b\u306e\u306f\u8ca0\u8377\u5bb9\u91cf\u304c\u4f4e\u3044\u914d\u7dda\u3092\u4f7f\u3063\u3066\u9060\u56de\u308a\u306b\u306a\u3063\u305f\u3001\u3068\u3044\u3046\u306e\u304c\u3088\u304f\u308f\u304b\u3089\u306a\u3044\u3002\u5bb9\u91cf\u304c\u4f4e\u3044\u306e\u306f\u77ed\u3044\u306e\u3067\u306f (\u4e95\u53e3\u5148\u751f@JAIST)? \u2192 \u8fc2\u56de\u3088\u308a\u3082 long line \u3092\u524a\u3063\u3061\u3083\u3063\u305f\u3053\u3068\u304c\u554f\u984c\u3002
\n\u98ef\u7530\u5148\u751f: \u898f\u5247\u7684\u306b\u305f\u304f\u3055\u3093\u5165\u308c\u308b\u3088\u308a\u3001\u30e9\u30f3\u30c0\u30e0\u306b\u5c11\u306a\u304f\u3044\u308c\u305f\u307b\u3046\u304c\u7bc0\u96fb\u306b\u306a\u3063\u3066\u3044\u3044\u3002\u914d\u7dda\u306f2\u5c64\u306b\u5236\u9650\u3057\u3066\u3044\u308b\u3002\u96fb\u529b\u306e\u524a\u6e1b\u3092\u76ee\u7684\u306b frequency \u304c\u9ad8\u3044\u306e\u3092 SW line \u306b\u5272\u308a\u5f53\u3066\u3066\u3057\u307e\u3046\u306e\u3067\u3001\u914d\u7dda\u304c\u9577\u3044\u3084\u3064\u304c SW line \u306b\u884c\u3051\u305a\u306b\u9045\u304f\u306a\u3063\u305f\u308a\u3057\u3066\u3044\u308b\u304b\u3082\u3057\u308c\u306a\u3044\u3002
\n[ \u9ad8\u3044\u30b9\u30eb\u30fc\u30d7\u30c3\u30c8\u3092\u5b9f\u73fe\u3059\u308b\u7d44\u307f\u5408\u308f\u305b\u751f\u6210\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u306e\u63d0\u6848\u3068\u5b9f\u88c5 ]
\n[ FPGA\u3092\u7528\u3044\u305fHMMER\u306e\u9ad8\u901f\u5316 ]
\n\u8ab0\u304c\u4f7f\u3046\u306e\u304b? \u3068\u3044\u3046\u6839\u672c\u7684\u306a\u8cea\u554f\u304c\u5b89\u6c38\u5148\u751f\u304b\u3089\u3002
\n\u3067\u3082\u3001\u6b21\u4e16\u4ee3\u30b7\u30fc\u30b1\u30f3\u30b5\u3068\u304b\u3067\u914d\u5217\u304c\u3069\u3093\u3069\u3093\u51fa\u3066\u304f\u308b\u306e\u3067\u3001\u8a08\u7b97\u6a5f\u306e\u30b3\u30b9\u30c8\u3068\u6027\u80fd\u3082\u306a\u3093\u3068\u304b\u3057\u306a\u3051\u308c\u3070\u3044\u3051\u306a\u3044\u6642\u4ee3\u304b\u306a\u3001\u3068\u601d\u3044\u307e\u3059\u3002
\n[ FPGA\u3092\u7528\u3044\u305f\u6c4e\u7528\u751f\u5316\u5b66\u30b7\u30df\u30e5\u30ec\u30fc\u30bf\u306b\u304a\u3051\u308b\u30cf\u30fc\u30c9\u30a6\u30a7\u30a2\u30e2\u30b8\u30e5\u30fc\u30eb\u81ea\u52d5\u751f\u6210\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u306e\u5b9f\u6a5f\u8a55\u4fa1 ]
\n\u6f14\u7b97\u5668\u306e\u30b9\u30b1\u30b8\u30e5\u30fc\u30ea\u30f3\u30b0\u306f\u3069\u3046\u306a\u3063\u3066\u308b\u306e (\u5f18\u4e2d\u5148\u751f)\uff1f : DFG \u3092\u4f5c\u3063\u305f\u5f8c\u30d1\u30a4\u30d7\u30e9\u30a4\u30f3\u3092\u4f5c\u308b\u6642\u70b9\u3067\u3084\u3063\u3066\u307e\u3059 (\u3057\u3070\u305f\u3055\u3093)
\n\u7d50\u5408\u3057\u305f\u307b\u3046\u304c\u5468\u6ce2\u6570\u304c\u3042\u304c\u308b\u306e\u306f\u306a\u305c (\u540d\u53e4\u5c4b\u5148\u751f) ? : \u305d\u3046\u3044\u3046\u3053\u3068\u3082\u3042\u308a\u307e\u3059…
\n[ \u30a4\u30f3\u30bf\u30e9\u30af\u30c6\u30a3\u30d6\u30fb\u30b9\u30fc\u30d1\u30fc\u30b3\u30f3\u30d4\u30e5\u30fc\u30c6\u30a3\u30f3\u30b0\u74b0\u5883\u306e\u5b9f\u73fe\u3078\u5411\u3051\u3066 ]
\n\u53ef\u8996\u5316: spatial resolution \u306f\u9ad8\u3044\u304c\u3001\u9045\u5ef6\u306b\u306f\u5bdb\u5bb9
\n\u53ef\u89e6\u5316: spatial resolution \u306f\u4f4e\u3044\u304c\u3001\u9045\u5ef6\u306b\u306f\u3082\u306e\u3059\u3054\u304f\u654f\u611f
\nSimulation caching: remote \u3067\u3084\u3063\u3066\u3044\u308b\u9ad8\u89e3\u50cf\u5ea6\u306e\u8a08\u7b97\u3068\u3001local \u3067\u3084\u3063\u3066\u3044\u308b\u4f4e\u89e3\u50cf\u5ea6\u306e\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3092\u5354\u8abf\u3055\u305b\u308b\u3002\u529b\u899a\u306e\u307b\u3046\u306f local \u306e\u3092\u30ea\u30a2\u30eb\u30bf\u30a4\u30e0\u3067\u3001\u8996\u899a\u306e\u307b\u3046\u306f remote \u306e\u3092\u30c7\u30fc\u30bf\u304c\u5c4a\u304d\u6b21\u7b2c\u5448\u793a\u3002\u3067\u3082\u3001\u300c\u5c4a\u304d\u6b21\u7b2c\u300d\u3060\u3068\u3001\u30b8\u30c3\u30bf\u304c\u3042\u308b\u306e\u3067\u3001\u3069\u3046\u3057\u3088\u3046\u3002\u3053\u3053\u306f\u4eba\u306b\u3088\u3063\u3066\u597d\u307f\u304c\u5927\u304d\u304f\u7570\u306a\u308b\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"
[ MeP \u306e\u62e1\u5f35\u6a5f\u80fd\u3092\u7528\u3044\u305f\u518d\u69cb\u6210\u578b\u30d7\u30ed\u30bb\u30c3\u30b5Hy-DiSC\u306e\u6027\u80fd\u8a55\u4fa1 ] MeP \u3067\u9010\u6b21\u51e6\u7406\u3001DS-HIE \u30d7\u30ed\u30bb\u30c3\u30b5\u3067\u30b9\u30c8\u30ea\u30fc\u30df\u30f3\u30b0\u51e6\u7406\u3002 \u30c7\u30fc\u30bf\u4f9b\u7d66\u306a\u3069\u306f MeP \u304c\u884c\u3046\u306e\u3067\u3001\u30ad\u30e3\u30c3\u30b7\u30e5\u30df\u30b9\u304c\u6027\u80fd\u4f4e\u4e0b\u306e\u8981\u56e0\u306b\u306a\u308b … “RECONF May.14, 2009” \u306e<\/span>\u7d9a\u304d\u3092\u8aad\u3080<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"activitypub_content_warning":"","activitypub_content_visibility":"","activitypub_max_image_attachments":4,"activitypub_interaction_policy_quote":"anyone","activitypub_status":"","footnotes":""},"categories":[10],"tags":[],"class_list":["post-2368","post","type-post","status-publish","format-standard","hentry","category-conference-logs"],"_links":{"self":[{"href":"https:\/\/yasu2.prosou.nu\/blog\/index.php\/wp-json\/wp\/v2\/posts\/2368","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/yasu2.prosou.nu\/blog\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/yasu2.prosou.nu\/blog\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/yasu2.prosou.nu\/blog\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/yasu2.prosou.nu\/blog\/index.php\/wp-json\/wp\/v2\/comments?post=2368"}],"version-history":[{"count":0,"href":"https:\/\/yasu2.prosou.nu\/blog\/index.php\/wp-json\/wp\/v2\/posts\/2368\/revisions"}],"wp:attachment":[{"href":"https:\/\/yasu2.prosou.nu\/blog\/index.php\/wp-json\/wp\/v2\/media?parent=2368"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/yasu2.prosou.nu\/blog\/index.php\/wp-json\/wp\/v2\/categories?post=2368"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/yasu2.prosou.nu\/blog\/index.php\/wp-json\/wp\/v2\/tags?post=2368"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}