-
Notifications
You must be signed in to change notification settings - Fork 96
AMD Ryzen Threadripper 2950X 16 Core Processor
David Huang edited this page Mar 7, 2019
·
2 revisions
AMD Ryzen Threadripper 2950X @ 4.4 GHz
4×16GB DDR4 3000MHz 15-17-17-36
$ lscpu
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 32
On-line CPU(s) list: 0-31
Thread(s) per core: 2
Core(s) per socket: 16
Socket(s): 1
NUMA node(s): 2
Vendor ID: AuthenticAMD
CPU family: 23
Model: 8
Model name: AMD Ryzen Threadripper 2950X 16-Core Processor
Stepping: 2
CPU MHz: 1884.334
CPU max MHz: 3500.0000
CPU min MHz: 2200.0000
BogoMIPS: 6985.91
Virtualization: AMD-V
L1d cache: 32K
L1i cache: 64K
L2 cache: 512K
L3 cache: 8192K
NUMA node0 CPU(s): 0-7,16-23
NUMA node1 CPU(s): 8-15,24-31
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid amd_dcm aperfmperf pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb hw_pstate sme ssbd sev ibpb vmmcall fsgsbase bmi1 avx2 smep bmi2 rdseed adx smap clflushopt sha_ni xsaveopt xsavec xgetbv1 xsaves clzero irperf xsaveerptr arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif overflow_recov succor smca
$ numactl --hardware
available: 2 nodes (0-1)
node 0 cpus: 0 1 2 3 4 5 6 7 16 17 18 19 20 21 22 23
node 0 size: 32119 MB
node 0 free: 28327 MB
node 1 cpus: 8 9 10 11 12 13 14 15 24 25 26 27 28 29 30 31
node 1 size: 32223 MB
node 1 free: 27259 MB
node distances:
node 0 1
0: 10 16
1: 16 10
$ numactl --cpunodebind 1 --membind 1 ./tinymembench
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 : 9447.3 MB/s
C copy backwards (32 byte blocks) : 9445.9 MB/s
C copy backwards (64 byte blocks) : 9434.8 MB/s
C copy : 9517.7 MB/s
C copy prefetched (32 bytes step) : 10055.4 MB/s
C copy prefetched (64 bytes step) : 10034.2 MB/s
C 2-pass copy : 8102.9 MB/s
C 2-pass copy prefetched (32 bytes step) : 9065.7 MB/s
C 2-pass copy prefetched (64 bytes step) : 9092.5 MB/s
C fill : 11019.8 MB/s
C fill (shuffle within 16 byte blocks) : 11018.5 MB/s
C fill (shuffle within 32 byte blocks) : 11020.2 MB/s
C fill (shuffle within 64 byte blocks) : 11030.3 MB/s
---
standard memcpy : 17497.6 MB/s
standard memset : 14133.8 MB/s (0.2%)
---
MOVSB copy : 10706.6 MB/s
MOVSD copy : 10718.4 MB/s
SSE2 copy : 10566.6 MB/s
SSE2 nontemporal copy : 17281.6 MB/s
SSE2 copy prefetched (32 bytes step) : 10347.9 MB/s
SSE2 copy prefetched (64 bytes step) : 10468.8 MB/s
SSE2 nontemporal copy prefetched (32 bytes step) : 17839.4 MB/s
SSE2 nontemporal copy prefetched (64 bytes step) : 17826.5 MB/s
SSE2 2-pass copy : 9284.6 MB/s
SSE2 2-pass copy prefetched (32 bytes step) : 9752.5 MB/s
SSE2 2-pass copy prefetched (64 bytes step) : 9764.5 MB/s
SSE2 2-pass nontemporal copy : 5718.8 MB/s
SSE2 fill : 13713.9 MB/s
SSE2 nontemporal fill : 43439.4 MB/s
==========================================================================
== 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 : 1.3 ns / 1.6 ns
262144 : 1.6 ns / 1.8 ns
524288 : 3.2 ns / 4.1 ns
1048576 : 5.8 ns / 7.3 ns
2097152 : 7.4 ns / 8.6 ns
4194304 : 8.4 ns / 9.1 ns
8388608 : 17.9 ns / 25.0 ns
16777216 : 44.5 ns / 61.5 ns
33554432 : 61.5 ns / 75.7 ns
67108864 : 70.8 ns / 81.2 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 : 1.6 ns / 1.7 ns
524288 : 1.8 ns / 1.9 ns
1048576 : 4.5 ns / 5.9 ns
2097152 : 5.8 ns / 7.0 ns
4194304 : 6.6 ns / 7.3 ns
8388608 : 10.2 ns / 13.8 ns
16777216 : 38.1 ns / 54.3 ns
33554432 : 53.3 ns / 67.0 ns
67108864 : 61.2 ns / 70.7 ns
$ numactl --cpunodebind 0 --membind 1 ./tinymembench
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 : 7250.6 MB/s
C copy backwards (32 byte blocks) : 7310.5 MB/s
C copy backwards (64 byte blocks) : 7274.0 MB/s
C copy : 7394.4 MB/s
C copy prefetched (32 bytes step) : 7729.0 MB/s
C copy prefetched (64 bytes step) : 7698.6 MB/s
C 2-pass copy : 6528.7 MB/s
C 2-pass copy prefetched (32 bytes step) : 6890.0 MB/s (0.1%)
C 2-pass copy prefetched (64 bytes step) : 6907.3 MB/s
C fill : 9125.4 MB/s
C fill (shuffle within 16 byte blocks) : 9130.7 MB/s
C fill (shuffle within 32 byte blocks) : 9133.6 MB/s
C fill (shuffle within 64 byte blocks) : 9143.7 MB/s
---
standard memcpy : 10898.2 MB/s
standard memset : 10705.2 MB/s (0.1%)
---
MOVSB copy : 6941.6 MB/s
MOVSD copy : 6937.7 MB/s
SSE2 copy : 7723.1 MB/s (0.2%)
SSE2 nontemporal copy : 11045.3 MB/s
SSE2 copy prefetched (32 bytes step) : 7332.2 MB/s
SSE2 copy prefetched (64 bytes step) : 7328.5 MB/s
SSE2 nontemporal copy prefetched (32 bytes step) : 11130.3 MB/s
SSE2 nontemporal copy prefetched (64 bytes step) : 11110.4 MB/s
SSE2 2-pass copy : 7000.6 MB/s
SSE2 2-pass copy prefetched (32 bytes step) : 7329.4 MB/s
SSE2 2-pass copy prefetched (64 bytes step) : 7346.4 MB/s
SSE2 2-pass nontemporal copy : 3389.2 MB/s
SSE2 fill : 10438.6 MB/s
SSE2 nontemporal fill : 15676.1 MB/s
==========================================================================
== 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 : 1.6 ns / 1.8 ns
524288 : 3.3 ns / 4.2 ns
1048576 : 5.9 ns / 7.3 ns
2097152 : 7.6 ns / 8.6 ns
4194304 : 8.4 ns / 9.1 ns
8388608 : 23.7 ns / 35.5 ns
16777216 : 68.9 ns / 98.4 ns
33554432 : 97.2 ns / 122.7 ns
67108864 : 109.6 ns / 131.4 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 : 1.6 ns / 1.8 ns
524288 : 1.8 ns / 1.9 ns
1048576 : 4.5 ns / 5.9 ns
2097152 : 6.0 ns / 7.0 ns
4194304 : 6.6 ns / 7.3 ns
8388608 : 13.4 ns / 18.6 ns
16777216 : 62.4 ns / 91.3 ns
33554432 : 89.2 ns / 114.3 ns
67108864 : 99.9 ns / 121.4 ns
Kernel 4.9.140-tegra #1 SMP PREEMPT Wed Mar 13 00:32:22 PDT 2019 aarch64 GNU/Linux Under xorg, no compositor active, no browser or other cpu hogs.
tinymembench v0.4.9 (simple benchmark for memory thr
==========================================================================
== 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 : 2949.7 MB/s (3.8%)
C copy backwards (32 byte blocks) : 3011.8 MB/s
C copy backwards (64 byte blocks) : 3029.2 MB/s
C copy : 3642.2 MB/s (4.1%)
C copy prefetched (32 bytes step) : 3824.4 MB/s (0.3%)
C copy prefetched (64 bytes step) : 3825.3 MB/s (0.4%)
C 2-pass copy : 2726.2 MB/s
C 2-pass copy prefetched (32 bytes step) : 2902.6 MB/s (2.5%)
C 2-pass copy prefetched (64 bytes step) : 2928.3 MB/s (0.3%)
C fill : 8541.0 MB/s (0.2%)
C fill (shuffle within 16 byte blocks) : 8518.5 MB/s (2.1%)
C fill (shuffle within 32 byte blocks) : 8537.1 MB/s (0.1%)
C fill (shuffle within 64 byte blocks) : 8528.7 MB/s (0.2%)
---
standard memcpy : 3558.8 MB/s
standard memset : 8520.2 MB/s
---
NEON LDP/STP copy : 3633.9 MB/s (4.2%)
NEON LDP/STP copy pldl2strm (32 bytes step) : 1451.0 MB/s (0.3%)
NEON LDP/STP copy pldl2strm (64 bytes step) : 1450.9 MB/s (0.5%)
NEON LDP/STP copy pldl1keep (32 bytes step) : 3882.5 MB/s (3.9%)
NEON LDP/STP copy pldl1keep (64 bytes step) : 3884.0 MB/s (0.4%)
NEON LD1/ST1 copy : 3630.8 MB/s (0.3%)
NEON STP fill : 8537.8 MB/s
NEON STNP fill : 8544.9 MB/s (1.7%)
ARM LDP/STP copy : 3635.8 MB/s (0.3%)
ARM STP fill : 8544.8 MB/s (0.1%)
ARM STNP fill : 8549.2 MB/s (1.0%)
==========================================================================
== Framebuffer read tests. ==
== ==
== Many ARM devices use a part of the system memory as the framebuffer, ==
== typically mapped as uncached but with write-combining enabled. ==
== Writes to such framebuffers are quite fast, but reads are much ==
== slower and very sensitive to the alignment and the selection of ==
== CPU instructions which are used for accessing memory. ==
== ==
== Many x86 systems allocate the framebuffer in the GPU memory, ==
== accessible for the CPU via a relatively slow PCI-E bus. Moreover, ==
== PCI-E is asymmetric and handles reads a lot worse than writes. ==
== ==
== If uncached framebuffer reads are reasonably fast (at least 100 MB/s ==
== or preferably >300 MB/s), then using the shadow framebuffer layer ==
== is not necessary in Xorg DDX drivers, resulting in a nice overall ==
== performance improvement. For example, the xf86-video-fbturbo DDX ==
== uses this trick. ==
==========================================================================
NEON LDP/STP copy (from framebuffer) : 766.0 MB/s
NEON LDP/STP 2-pass copy (from framebuffer) : 688.8 MB/s
NEON LD1/ST1 copy (from framebuffer) : 770.6 MB/s (0.1%)
NEON LD1/ST1 2-pass copy (from framebuffer) : 681.3 MB/s (0.3%)
ARM LDP/STP copy (from framebuffer) : 766.1 MB/s
ARM LDP/STP 2-pass copy (from framebuffer) : 689.1 MB/s
==========================================================================
== 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 / 0.1 ns
2048 : 0.0 ns / 0.1 ns
4096 : 0.0 ns / 0.1 ns
8192 : 0.0 ns / 0.1 ns
16384 : 0.1 ns / 0.1 ns
32768 : 1.7 ns / 2.9 ns
65536 : 6.4 ns / 9.5 ns
131072 : 9.6 ns / 12.3 ns
262144 : 13.7 ns / 17.0 ns
524288 : 15.8 ns / 19.7 ns
1048576 : 17.3 ns / 22.1 ns
2097152 : 42.1 ns / 64.2 ns
4194304 : 98.5 ns / 138.1 ns
8388608 : 143.9 ns / 186.3 ns
16777216 : 167.2 ns / 211.2 ns
33554432 : 180.1 ns / 227.1 ns
67108864 : 200.0 ns / 260.2 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 : 6.4 ns / 9.4 ns
131072 : 9.5 ns / 12.2 ns
262144 : 11.2 ns / 13.1 ns
524288 : 12.1 ns / 13.5 ns
1048576 : 12.8 ns / 13.6 ns
2097152 : 27.0 ns / 33.0 ns
4194304 : 90.6 ns / 127.8 ns
8388608 : 123.9 ns / 153.8 ns
16777216 : 139.5 ns / 161.2 ns
33554432 : 147.2 ns / 163.6 ns
67108864 : 154.0 ns / 167.6 ns