AMD INVLPGB Linux Patch Performance Benchmarks for a future article looking at AMD broadcast TLB invalidation Linux kernel patches with the INVLPGB instruction on newer AMD Zen 3 processors. Linux 6.13 Git: Processor: AMD EPYC 9655P 96-Core @ 2.60GHz (96 Cores / 192 Threads), Motherboard: Supermicro Super Server H13SSL-N v1.01 (3.0 BIOS), Chipset: AMD 1Ah, Memory: 12 x 64GB DDR5-6000MT/s Micron MTC40F2046S1RC64BDY QSFF, Disk: 3201GB Micron_7450_MTFDKCB3T2TFS + 257GB Flash Drive, Graphics: ASPEED, Network: 2 x Broadcom NetXtreme BCM5720 PCIe OS: Ubuntu 24.10, Kernel: 6.13.0-rc4-phx-stock (x86_64), Desktop: GNOME Shell 47.0, Display Server: X Server, Compiler: GCC 14.2.0, File-System: ext4, Screen Resolution: 1024x768 INVLPGB Patched: Processor: AMD EPYC 9655P 96-Core @ 2.60GHz (96 Cores / 192 Threads), Motherboard: Supermicro Super Server H13SSL-N v1.01 (3.0 BIOS), Chipset: AMD 1Ah, Memory: 12 x 64GB DDR5-6000MT/s Micron MTC40F2046S1RC64BDY QSFF, Disk: 3201GB Micron_7450_MTFDKCB3T2TFS + 257GB Flash Drive, Graphics: ASPEED, Network: 2 x Broadcom NetXtreme BCM5720 PCIe OS: Ubuntu 24.10, Kernel: 6.13.0-rc4-phx-broadcast-tlb (x86_64), Desktop: GNOME Shell 47.0, Display Server: X Server, Compiler: GCC 14.2.0, File-System: ext4, Screen Resolution: 1024x768 Apache Cassandra 5.0 Test: Writes Op/s > Higher Is Better Linux 6.13 Git .. 438419 |=================================================== INVLPGB Patched . 452025 |===================================================== Apache IoTDB 1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 100 point/sec > Higher Is Better Linux 6.13 Git .. 96806809 |=================================================== INVLPGB Patched . 97751377 |=================================================== Apache IoTDB 1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 100 Average Latency < Lower Is Better Linux 6.13 Git .. 47.85 |====================================================== INVLPGB Patched . 47.03 |===================================================== Apache IoTDB 1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400 point/sec > Higher Is Better Linux 6.13 Git .. 99520170 |================================================== INVLPGB Patched . 100250585 |================================================== Apache IoTDB 1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400 Average Latency < Lower Is Better Linux 6.13 Git .. 168.36 |===================================================== INVLPGB Patched . 167.95 |===================================================== Apache IoTDB 1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 100 point/sec > Higher Is Better Linux 6.13 Git .. 120738449 |================================================== INVLPGB Patched . 121681953 |================================================== Apache IoTDB 1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 100 Average Latency < Lower Is Better Linux 6.13 Git .. 61.49 |====================================================== INVLPGB Patched . 60.93 |====================================================== Apache IoTDB 1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400 point/sec > Higher Is Better Linux 6.13 Git .. 125423055 |================================================== INVLPGB Patched . 124631354 |================================================== Apache IoTDB 1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400 Average Latency < Lower Is Better Linux 6.13 Git .. 211.81 |==================================================== INVLPGB Patched . 214.48 |===================================================== Apache IoTDB 1.2 Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 100 point/sec > Higher Is Better Linux 6.13 Git .. 117887210 |================================================== INVLPGB Patched . 118890170 |================================================== Apache IoTDB 1.2 Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 100 Average Latency < Lower Is Better Linux 6.13 Git .. 40.40 |====================================================== INVLPGB Patched . 40.04 |====================================================== Apache IoTDB 1.2 Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400 point/sec > Higher Is Better Linux 6.13 Git .. 121237884 |================================================== INVLPGB Patched . 120726010 |================================================== Apache IoTDB 1.2 Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400 point/sec > Higher Is Better Linux 6.13 Git .. 141171394 |================================================= INVLPGB Patched . 144092202 |================================================== Apache IoTDB 1.2 Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400 Average Latency < Lower Is Better Linux 6.13 Git .. 212.98 |===================================================== INVLPGB Patched . 206.65 |=================================================== Blender 4.3 Blend File: Barbershop - Compute: CPU-Only Seconds < Lower Is Better Linux 6.13 Git .. 124.03 |===================================================== INVLPGB Patched . 124.48 |===================================================== ClickHouse 22.12.3.5 100M Rows Hits Dataset, First Run / Cold Cache Queries Per Minute, Geo Mean > Higher Is Better Linux 6.13 Git .. 738.99 |===================================================== INVLPGB Patched . 735.32 |===================================================== ClickHouse 22.12.3.5 100M Rows Hits Dataset, Second Run Queries Per Minute, Geo Mean > Higher Is Better Linux 6.13 Git .. 764.71 |===================================================== INVLPGB Patched . 749.62 |==================================================== ClickHouse 22.12.3.5 100M Rows Hits Dataset, Third Run Queries Per Minute, Geo Mean > Higher Is Better Linux 6.13 Git .. 751.84 |==================================================== INVLPGB Patched . 759.44 |===================================================== DaCapo Benchmark 23.11 Java Test: Jython msec < Lower Is Better Linux 6.13 Git .. 4320 |======================================================= INVLPGB Patched . 4158 |===================================================== DaCapo Benchmark 23.11 Java Test: GraphChi msec < Lower Is Better Linux 6.13 Git .. 2363 |======================================================= INVLPGB Patched . 2323 |====================================================== DaCapo Benchmark 23.11 Java Test: Apache Kafka msec < Lower Is Better Linux 6.13 Git .. 5046 |======================================================= INVLPGB Patched . 5050 |======================================================= DaCapo Benchmark 23.11 Java Test: jMonkeyEngine msec < Lower Is Better Linux 6.13 Git .. 6808 |======================================================= INVLPGB Patched . 6806 |======================================================= DaCapo Benchmark 23.11 Java Test: Apache Xalan XSLT msec < Lower Is Better Linux 6.13 Git .. 816 |======================================================== INVLPGB Patched . 812 |======================================================== DaCapo Benchmark 23.11 Java Test: Avrora AVR Simulation Framework msec < Lower Is Better Linux 6.13 Git .. 2585 |======================================================= INVLPGB Patched . 2556 |====================================================== DaCapo Benchmark 23.11 Java Test: BioJava Biological Data Framework msec < Lower Is Better Linux 6.13 Git .. 4519 |======================================================= INVLPGB Patched . 4547 |======================================================= DaCapo Benchmark 23.11 Java Test: Zxing 1D/2D Barcode Image Processing msec < Lower Is Better Linux 6.13 Git .. 648 |======================================================== INVLPGB Patched . 619 |===================================================== MariaDB 11.5 Test: oltp_read_only - Threads: 256 Queries Per Second > Higher Is Better Linux 6.13 Git .. 37688 |====================================================== INVLPGB Patched . 37868 |====================================================== MariaDB 11.5 Test: oltp_read_write - Threads: 256 Queries Per Second > Higher Is Better Linux 6.13 Git .. 172615 |===================================================== INVLPGB Patched . 173765 |===================================================== MariaDB 11.5 Test: oltp_update_index - Threads: 256 Queries Per Second > Higher Is Better Linux 6.13 Git .. 113658 |=================================================== INVLPGB Patched . 117679 |===================================================== Memcached 1.6.19 Set To Get Ratio: 1:5 Ops/sec > Higher Is Better Linux 6.13 Git .. 3569817.66 |================================================= INVLPGB Patched . 3583145.46 |================================================= NAMD 3.0 Input: ATPase with 327,506 Atoms ns/day > Higher Is Better Linux 6.13 Git .. 12.22 |====================================================== INVLPGB Patched . 12.22 |====================================================== NAMD 3.0 Input: STMV with 1,066,628 Atoms ns/day > Higher Is Better Linux 6.13 Git .. 4.21510 |==================================================== INVLPGB Patched . 4.20973 |==================================================== NAS Parallel Benchmarks 3.4 Test / Class: BT.C Total Mop/s > Higher Is Better Linux 6.13 Git .. 353192.16 |================================================== INVLPGB Patched . 356244.91 |================================================== NAS Parallel Benchmarks 3.4 Test / Class: EP.C Total Mop/s > Higher Is Better Linux 6.13 Git .. 11180.11 |================================================== INVLPGB Patched . 11450.17 |=================================================== NAS Parallel Benchmarks 3.4 Test / Class: IS.D Total Mop/s > Higher Is Better Linux 6.13 Git .. 7001.42 |==================================================== INVLPGB Patched . 6984.65 |==================================================== NAS Parallel Benchmarks 3.4 Test / Class: LU.C Total Mop/s > Higher Is Better Linux 6.13 Git .. 312554.83 |================================================== INVLPGB Patched . 311722.61 |================================================== NAS Parallel Benchmarks 3.4 Test / Class: MG.C Total Mop/s > Higher Is Better Linux 6.13 Git .. 145177.79 |================================================= INVLPGB Patched . 147724.66 |================================================== NAS Parallel Benchmarks 3.4 Test / Class: SP.C Total Mop/s > Higher Is Better Linux 6.13 Git .. 154232.79 |================================================== INVLPGB Patched . 155286.47 |================================================== nginx 1.23.2 Connections: 500 Requests Per Second > Higher Is Better Linux 6.13 Git .. 503771.76 |================================================= INVLPGB Patched . 517615.54 |================================================== OpenFOAM 10 Input: drivaerFastback, Medium Mesh Size - Mesh Time Seconds < Lower Is Better Linux 6.13 Git .. 109.86 |===================================================== INVLPGB Patched . 106.21 |=================================================== Quicksilver 20230818 Input: CORAL2 P1 Figure Of Merit > Higher Is Better Linux 6.13 Git .. 34086667 |================================================== INVLPGB Patched . 34996667 |=================================================== Quicksilver 20230818 Input: CORAL2 P2 Figure Of Merit > Higher Is Better Linux 6.13 Git .. 22056667 |=================================================== INVLPGB Patched . 22056667 |=================================================== RELION 5.0 Test: Basic - Device: CPU Seconds < Lower Is Better Linux 6.13 Git .. 214.44 |===================================================== INVLPGB Patched . 211.38 |==================================================== Renaissance 0.16 Test: ALS Movie Lens ms < Lower Is Better Linux 6.13 Git .. 18214.8 |==================================================== INVLPGB Patched . 18368.8 |==================================================== Renaissance 0.16 Test: Apache Spark Bayes ms < Lower Is Better Linux 6.13 Git .. 180.1 |====================================================== INVLPGB Patched . 181.7 |====================================================== Renaissance 0.16 Test: Savina Reactors.IO ms < Lower Is Better Linux 6.13 Git .. 5722.9 |===================================================== INVLPGB Patched . 5651.4 |==================================================== Renaissance 0.16 Test: Apache Spark PageRank ms < Lower Is Better Linux 6.13 Git .. 2261.3 |==================================================== INVLPGB Patched . 2283.3 |===================================================== Renaissance 0.16 Test: In-Memory Database Shootout ms < Lower Is Better Linux 6.13 Git .. 4511.6 |===================================================== INVLPGB Patched . 4512.9 |===================================================== RocksDB 9.0 Test: Read While Writing Op/s > Higher Is Better Linux 6.13 Git .. 14641889 |=================================================== INVLPGB Patched . 14737550 |=================================================== Rodinia 3.1 Test: OpenMP Leukocyte Seconds < Lower Is Better Linux 6.13 Git .. 31.67 |====================================================== INVLPGB Patched . 31.30 |===================================================== Speedb 2.7 Test: Random Read Op/s > Higher Is Better Linux 6.13 Git .. 622084083 |================================================== INVLPGB Patched . 627286939 |================================================== Speedb 2.7 Test: Update Random Op/s > Higher Is Better Linux 6.13 Git .. 519867 |===================================================== INVLPGB Patched . 522238 |===================================================== Speedb 2.7 Test: Read Random Write Random Op/s > Higher Is Better Linux 6.13 Git .. 3054826 |=================================================== INVLPGB Patched . 3106319 |==================================================== Timed CPython Compilation 3.10.6 Build Configuration: Default Seconds < Lower Is Better Linux 6.13 Git .. 13.07 |====================================================== INVLPGB Patched . 12.98 |====================================================== Timed CPython Compilation 3.10.6 Build Configuration: Released Build, PGO + LTO Optimized Seconds < Lower Is Better Linux 6.13 Git .. 190.91 |===================================================== INVLPGB Patched . 191.97 |===================================================== Timed Godot Game Engine Compilation 4.0 Time To Compile Seconds < Lower Is Better Linux 6.13 Git .. 80.69 |====================================================== INVLPGB Patched . 80.83 |====================================================== Timed LLVM Compilation 16.0 Build System: Ninja Seconds < Lower Is Better Linux 6.13 Git .. 90.42 |====================================================== INVLPGB Patched . 90.78 |====================================================== Timed LLVM Compilation 16.0 Build System: Unix Makefiles Seconds < Lower Is Better Linux 6.13 Git .. 160.67 |===================================================== INVLPGB Patched . 161.06 |===================================================== Whisper.cpp 1.6.2 Model: ggml-small.en - Input: 2016 State of the Union Seconds < Lower Is Better Linux 6.13 Git .. 225.42 |===================================================== INVLPGB Patched . 219.87 |==================================================== Whisper.cpp 1.6.2 Model: ggml-medium.en - Input: 2016 State of the Union Seconds < Lower Is Better Linux 6.13 Git .. 460.62 |===================================================== INVLPGB Patched . 453.07 |==================================================== Xmrig 6.21 Variant: GhostRider - Hash Count: 1M H/s > Higher Is Better Linux 6.13 Git .. 13706.9 |============================================ INVLPGB Patched . 16061.4 |====================================================