tinymembench v0.4.9 (simple benchmark for memory throughput and latency) ========================================================================== == Memory bandwidth tests == == == == Note 1: 1MB = 1000000 bytes == == Note 2: Results for 'copy' tests show how many bytes can be == == copied per second (adding together read and writen == == bytes would have provided twice higher numbers) == == Note 3: 2-pass copy means that we are using a small temporary buffer == == to first fetch data into it, and only then write it to the == == destination (source -> L1 cache, L1 cache -> destination) == == Note 4: If sample standard deviation exceeds 0.1%, it is shown in == == brackets == ========================================================================== C copy backwards : 6026.5 MB/s (1.7%) C copy backwards (32 byte blocks) : 6492.3 MB/s (4.0%) C copy backwards (64 byte blocks) : 6510.8 MB/s (1.4%) C copy : 6536.8 MB/s (1.3%) C copy prefetched (32 bytes step) : 6682.7 MB/s (1.2%) C copy prefetched (64 bytes step) : 6691.9 MB/s (1.6%) C 2-pass copy : 5628.0 MB/s (1.2%) C 2-pass copy prefetched (32 bytes step) : 6033.4 MB/s (2.1%) C 2-pass copy prefetched (64 bytes step) : 5932.2 MB/s (2.3%) C fill : 7789.1 MB/s (1.1%) C fill (shuffle within 16 byte blocks) : 7786.7 MB/s (1.2%) C fill (shuffle within 32 byte blocks) : 7776.3 MB/s (1.6%) C fill (shuffle within 64 byte blocks) : 7781.6 MB/s (0.9%) --- standard memcpy : 11433.8 MB/s (1.0%) standard memset : 33431.6 MB/s (1.9%) --- MOVSB copy : 5597.5 MB/s (1.4%) MOVSD copy : 5596.8 MB/s (1.0%) SSE2 copy : 6803.1 MB/s (1.7%) SSE2 nontemporal copy : 11472.7 MB/s (1.2%) SSE2 copy prefetched (32 bytes step) : 6647.1 MB/s (1.3%) SSE2 copy prefetched (64 bytes step) : 6638.5 MB/s (1.9%) SSE2 nontemporal copy prefetched (32 bytes step) : 12191.6 MB/s (0.7%) SSE2 nontemporal copy prefetched (64 bytes step) : 12158.2 MB/s (1.1%) SSE2 2-pass copy : 6226.0 MB/s (1.4%) SSE2 2-pass copy prefetched (32 bytes step) : 6205.9 MB/s (1.5%) SSE2 2-pass copy prefetched (64 bytes step) : 6185.8 MB/s (1.3%) SSE2 2-pass nontemporal copy : 4283.9 MB/s (2.0%) SSE2 fill : 8767.7 MB/s (1.3%) SSE2 nontemporal fill : 33556.3 MB/s (2.4%) ========================================================================== == Memory latency test == == == == Average time is measured for random memory accesses in the buffers == == of different sizes. The larger is the buffer, the more significant == == are relative contributions of TLB, L1/L2 cache misses and SDRAM == == accesses. For extremely large buffer sizes we are expecting to see == == page table walk with several requests to SDRAM for almost every == == memory access (though 64MiB is not nearly large enough to experience == == this effect to its fullest). == == == == Note 1: All the numbers are representing extra time, which needs to == == be added to L1 cache latency. The cycle timings for L1 cache == == latency can be usually found in the processor documentation. == == Note 2: Dual random read means that we are simultaneously performing == == two independent memory accesses at a time. In the case if == == the memory subsystem can't handle multiple outstanding == == requests, dual random read has the same timings as two == == single reads performed one after another. == ========================================================================== block size : single random read / dual random read, [MADV_NOHUGEPAGE] 1024 : 0.0 ns / 0.0 ns 2048 : 0.0 ns / 0.0 ns 4096 : 0.0 ns / 0.0 ns 8192 : 0.0 ns / 0.0 ns 16384 : 0.0 ns / 0.0 ns 32768 : 0.0 ns / 0.0 ns 65536 : 0.9 ns / 1.3 ns 131072 : 1.4 ns / 1.7 ns 262144 : 1.7 ns / 1.8 ns 524288 : 2.8 ns / 3.4 ns 1048576 : 6.4 ns / 8.1 ns 2097152 : 8.2 ns / 9.4 ns 4194304 : 15.9 ns / 20.5 ns 8388608 : 56.2 ns / 79.7 ns 16777216 : 81.2 ns / 104.2 ns 33554432 : 92.2 ns / 112.3 ns 67108864 : 99.3 ns / 117.0 ns block size : single random read / dual random read, [MADV_HUGEPAGE] 1024 : 0.0 ns / 0.0 ns 2048 : 0.0 ns / 0.0 ns 4096 : 0.0 ns / 0.0 ns 8192 : 0.0 ns / 0.0 ns 16384 : 0.0 ns / 0.0 ns 32768 : 0.0 ns / 0.0 ns 65536 : 0.9 ns / 1.3 ns 131072 : 1.4 ns / 1.7 ns 262144 : 1.7 ns / 1.8 ns 524288 : 2.0 ns / 2.1 ns 1048576 : 5.2 ns / 6.6 ns 2097152 : 7.0 ns / 7.9 ns 4194304 : 10.5 ns / 13.1 ns 8388608 : 52.3 ns / 76.2 ns 16777216 : 75.1 ns / 93.4 ns 33554432 : 85.3 ns / 98.9 ns 67108864 : 93.2 ns / 108.5 ns tinymembench v0.4.9 (simple benchmark for memory throughput and latency)