Java AMD EPYC Tuning Performance

Java workload tuning on AMD EPYC Turin. Benchmarks by Michael Larabel for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2411302-NE-JAVAAMDEP26
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts
Allow Limiting Results To Certain Suite(s)

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs
Condense Test Profiles With Multiple Version Results Into Single Result Graphs

Table

Show Detailed System Result Table

Sensor Monitoring

Show Accumulated Sensor Monitoring Data For Displayed Results
Generate Power Efficiency / Performance Per Watt Results

Run Management

Highlight
Result
Toggle/Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
Stock
November 29
  1 Hour, 28 Minutes
Java Throughput
November 30
  2 Hours, 10 Minutes
Java Latency
November 30
  1 Hour, 49 Minutes
Power Efficiency
November 30
  2 Hours, 1 Minute
Invert Behavior (Only Show Selected Data)
  1 Hour, 52 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


Java AMD EPYC Tuning PerformanceOpenBenchmarking.orgPhoronix Test SuiteAMD EPYC 9575F 64-Core @ 3.30GHz (64 Cores / 128 Threads)Supermicro Super Server H13SSL-N v1.01 (3.0 BIOS)AMD 1Ah12 x 64GB DDR5-6000MT/s Micron MTC40F2046S1RC64BDY QSFF3201GB Micron_7450_MTFDKCB3T2TFSASPEED2 x Broadcom NetXtreme BCM5720 PCIeUbuntu 24.106.12.0-rc7-linux-pm-next-phx (x86_64)GNOME Shell 47.0X ServerGCC 14.2.0ext41024x768ProcessorMotherboardChipsetMemoryDiskGraphicsNetworkOSKernelDesktopDisplay ServerCompilerFile-SystemScreen ResolutionJava AMD EPYC Tuning Performance BenchmarksSystem Logs- Transparent Huge Pages: madvise- Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xb002116 - OpenJDK Runtime Environment (build 21.0.5+11-Ubuntu-1ubuntu124.10)- gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected

StockJava ThroughputJava LatencyPower EfficiencyResult OverviewPhoronix Test Suite100%113%126%140%153%Sunflow Rendering SystemApache CassandraRenaissanceDaCapo BenchmarkJava JMH

Java AMD EPYC Tuning Performancecassandra: Writesjava-jmh: Throughputrenaissance: Apache Spark Bayesrenaissance: Apache Spark PageRankrenaissance: Savina Reactors.IOrenaissance: Akka Unbalanced Cobwebbed Treerenaissance: Rand Forestrenaissance: In-Memory Database Shootoutrenaissance: Scala Dottyrenaissance: Finagle HTTP Requestsrenaissance: Genetic Algorithm Using Jenetics + Futuresrenaissance: ALS Movie Lensrenaissance: Gaussian Mixture Modeldacapobench: Eclipsedacapobench: Avrora AVR Simulation Frameworkdacapobench: Batik SVG Toolkitdacapobench: FOP Print Formatterdacapobench: Apache Tomcatdacapobench: Apache Xalan XSLTdacapobench: Apache Lucene Search Indexdacapobench: Apache Lucene Search Enginedacapobench: H2 Database Enginedacapobench: Jythondacapobench: Tradebeansdacapobench: Tradesoapdacapobench: BioJava Biological Data Frameworkdacapobench: GraphChidacapobench: jMonkeyEnginedacapobench: Apache Kafkadacapobench: Spring Bootdacapobench: Zxing 1D/2D Barcode Image Processingsunflow: Global Illumination + Image SynthesisStockJava ThroughputJava LatencyPower Efficiency464753312830489306.03151.92065.35172.110860.9383.54228.7448.82713.0894.015721.03254.3631522919253381037737224641402067376450643001404720946793506026415580.242459624313156984419.28152.82101.15166.810738.8391.83851.9458.82741.8777.515592.63182.4634322599253401069742223839592135376445132741404621796793507026365430.259491896314639272675.16152.02083.55081.410956.6382.93924.3457.52792.9903.615599.63194.4625822729333441057741225640532062376044132838402521266793506326615440.302356428315423179604.61193.62268.59954.213573.5606.65263.2655.23052.31291.720553.93925.39476303014695641368975344450722547488952403646530134406853508633116020.370OpenBenchmarking.org

Apache Cassandra

OpenBenchmarking.orgOp/s, More Is BetterApache Cassandra 5.0Test: WritesStockJava ThroughputJava LatencyPower Efficiency110K220K330K440K550KSE +/- 3783.12, N = 3SE +/- 784.82, N = 3SE +/- 2078.80, N = 3SE +/- 5044.93, N = 3464753459624491896356428

Java JMH

This very basic test profile runs the stock benchmark of the Java JMH benchmark via Maven. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/s, More Is BetterJava JMHThroughputStockJava ThroughputJava LatencyPower Efficiency70000M140000M210000M280000M350000M312830489306.03313156984419.28314639272675.16315423179604.61

Renaissance

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Apache Spark BayesStockJava ThroughputJava LatencyPower Efficiency4080120160200SE +/- 0.82, N = 3SE +/- 1.64, N = 15SE +/- 1.40, N = 15SE +/- 1.40, N = 3151.9152.8152.0193.6MIN: 128.87 / MAX: 209.2MIN: 127.3 / MAX: 382.17MIN: 121.4 / MAX: 361.56MIN: 157.46 / MAX: 378.44

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Apache Spark PageRankStockJava ThroughputJava LatencyPower Efficiency5001000150020002500SE +/- 8.00, N = 3SE +/- 15.93, N = 3SE +/- 18.35, N = 3SE +/- 19.39, N = 32065.32101.12083.52268.5MIN: 1460.75 / MAX: 2081.13MIN: 1452.23 / MAX: 2117.08MIN: 1424.26 / MAX: 2119.52MIN: 1959.01 / MAX: 2401.72

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Savina Reactors.IOStockJava ThroughputJava LatencyPower Efficiency2K4K6K8K10KSE +/- 54.20, N = 5SE +/- 37.22, N = 3SE +/- 49.91, N = 3SE +/- 125.38, N = 35172.15166.85081.49954.2MIN: 4973.43 / MAX: 8208.11MIN: 5095.51 / MAX: 7816.47MIN: 5006.45 / MAX: 8171.58MIN: 7679.87 / MAX: 10841.86

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Akka Unbalanced Cobwebbed TreeStockJava ThroughputJava LatencyPower Efficiency3K6K9K12K15KSE +/- 79.02, N = 3SE +/- 48.01, N = 3SE +/- 89.47, N = 3SE +/- 34.93, N = 310860.910738.810956.613573.5MIN: 10242.4 / MAX: 11077.04MIN: 10253.41 / MAX: 10941.76MIN: 10463.34 / MAX: 11370.49MIN: 12956.64 / MAX: 13724.14

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Random ForestStockJava ThroughputJava LatencyPower Efficiency130260390520650SE +/- 4.43, N = 3SE +/- 3.80, N = 15SE +/- 2.98, N = 10SE +/- 4.36, N = 15383.5391.8382.9606.6MIN: 347.9 / MAX: 457.62MIN: 342.67 / MAX: 493.98MIN: 339.12 / MAX: 447.53MIN: 536.2 / MAX: 741

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: In-Memory Database ShootoutStockJava ThroughputJava LatencyPower Efficiency11002200330044005500SE +/- 7.53, N = 3SE +/- 8.42, N = 3SE +/- 19.30, N = 3SE +/- 65.42, N = 34228.73851.93924.35263.2MIN: 4037.24 / MAX: 4846.07MIN: 3402.36 / MAX: 4284.31MIN: 3683.75 / MAX: 4557.34MIN: 4666.19 / MAX: 6061.56

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Scala DottyStockJava ThroughputJava LatencyPower Efficiency140280420560700SE +/- 8.07, N = 15SE +/- 6.53, N = 15SE +/- 4.40, N = 15SE +/- 20.51, N = 12448.8458.8457.5655.2MIN: 353.63 / MAX: 587.33MIN: 336.82 / MAX: 624.11MIN: 349.58 / MAX: 587.42MIN: 430.6 / MAX: 1104.36

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Finagle HTTP RequestsStockJava ThroughputJava LatencyPower Efficiency7001400210028003500SE +/- 20.00, N = 3SE +/- 7.56, N = 3SE +/- 20.65, N = 3SE +/- 35.51, N = 32713.02741.82792.93052.3MIN: 1897.78 / MAX: 2748.55MIN: 2037.17 / MAX: 2755.41MIN: 2055.2 / MAX: 2818.67MIN: 2255.4 / MAX: 3115.04

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Genetic Algorithm Using Jenetics + FuturesStockJava ThroughputJava LatencyPower Efficiency30060090012001500SE +/- 8.09, N = 3SE +/- 14.03, N = 15SE +/- 5.33, N = 3SE +/- 14.18, N = 3894.0777.5903.61291.7MIN: 841.31 / MAX: 918.73MIN: 667.17 / MAX: 1033.21MIN: 765.31 / MAX: 912.11MIN: 1207.53 / MAX: 1390.99

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: ALS Movie LensStockJava ThroughputJava LatencyPower Efficiency4K8K12K16K20KSE +/- 91.51, N = 3SE +/- 85.32, N = 3SE +/- 58.79, N = 3SE +/- 133.62, N = 315721.015592.615599.620553.9MIN: 15022.39 / MAX: 15859.07MIN: 15012.81 / MAX: 15741.43MIN: 15144.58 / MAX: 15795.77MIN: 19144.07 / MAX: 20820.8

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Gaussian Mixture ModelStockJava ThroughputJava LatencyPower Efficiency8001600240032004000SE +/- 24.37, N = 3SE +/- 9.38, N = 3SE +/- 10.95, N = 3SE +/- 30.43, N = 33254.33182.43194.43925.3MIN: 2372.34 / MAX: 3324.36MIN: 2333.74 / MAX: 3195.1MIN: 2370.4 / MAX: 3212.45MIN: 3849.24 / MAX: 4555.85

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance of various popular real-world Java workloads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: EclipseStockJava ThroughputJava LatencyPower Efficiency2K4K6K8K10KSE +/- 38.05, N = 3SE +/- 7.97, N = 3SE +/- 35.23, N = 3SE +/- 72.63, N = 106315634362589476

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: Avrora AVR Simulation FrameworkStockJava ThroughputJava LatencyPower Efficiency7001400210028003500SE +/- 23.08, N = 6SE +/- 17.33, N = 15SE +/- 14.51, N = 15SE +/- 27.41, N = 52291225922723030

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: Batik SVG ToolkitStockJava ThroughputJava LatencyPower Efficiency30060090012001500SE +/- 3.54, N = 8SE +/- 2.28, N = 8SE +/- 4.52, N = 8SE +/- 23.97, N = 129259259331469

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: FOP Print FormatterStockJava ThroughputJava LatencyPower Efficiency120240360480600SE +/- 1.47, N = 10SE +/- 1.32, N = 10SE +/- 2.24, N = 14SE +/- 6.80, N = 15338340344564

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: Apache TomcatStockJava ThroughputJava LatencyPower Efficiency30060090012001500SE +/- 1.76, N = 7SE +/- 8.41, N = 7SE +/- 3.16, N = 7SE +/- 7.83, N = 61037106910571368

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: Apache Xalan XSLTStockJava ThroughputJava LatencyPower Efficiency2004006008001000SE +/- 3.95, N = 10SE +/- 4.52, N = 9SE +/- 4.42, N = 9SE +/- 7.58, N = 8737742741975

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: Apache Lucene Search IndexStockJava ThroughputJava LatencyPower Efficiency7001400210028003500SE +/- 18.03, N = 3SE +/- 15.06, N = 3SE +/- 1.53, N = 3SE +/- 68.10, N = 152246223822563444

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: Apache Lucene Search EngineStockJava ThroughputJava LatencyPower Efficiency11002200330044005500SE +/- 19.75, N = 4SE +/- 44.71, N = 15SE +/- 48.20, N = 4SE +/- 61.68, N = 34140395940535072

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: H2 Database EngineStockJava ThroughputJava LatencyPower Efficiency5001000150020002500SE +/- 22.63, N = 5SE +/- 11.05, N = 3SE +/- 25.05, N = 4SE +/- 27.68, N = 52067213520622547

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: JythonStockJava ThroughputJava LatencyPower Efficiency10002000300040005000SE +/- 3.22, N = 4SE +/- 7.80, N = 4SE +/- 13.83, N = 4SE +/- 92.77, N = 123764376437604889

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: TradebeansStockJava ThroughputJava LatencyPower Efficiency11002200330044005500SE +/- 96.06, N = 15SE +/- 94.53, N = 15SE +/- 134.49, N = 12SE +/- 55.56, N = 155064451344135240

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: TradesoapStockJava ThroughputJava LatencyPower Efficiency8001600240032004000SE +/- 26.56, N = 4SE +/- 42.48, N = 15SE +/- 25.31, N = 15SE +/- 26.21, N = 153001274128383646

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: BioJava Biological Data FrameworkStockJava ThroughputJava LatencyPower Efficiency11002200330044005500SE +/- 7.35, N = 4SE +/- 12.85, N = 4SE +/- 14.73, N = 4SE +/- 84.18, N = 154047404640255301

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: GraphChiStockJava ThroughputJava LatencyPower Efficiency7001400210028003500SE +/- 8.25, N = 5SE +/- 22.52, N = 5SE +/- 14.98, N = 5SE +/- 23.13, N = 32094217921263440

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: jMonkeyEngineStockJava ThroughputJava LatencyPower Efficiency15003000450060007500SE +/- 1.00, N = 3SE +/- 1.33, N = 3SE +/- 0.33, N = 3SE +/- 1.15, N = 36793679367936853

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: Apache KafkaStockJava ThroughputJava LatencyPower Efficiency11002200330044005500SE +/- 0.58, N = 3SE +/- 5.51, N = 3SE +/- 5.84, N = 3SE +/- 5.81, N = 35060507050635086

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: Spring BootStockJava ThroughputJava LatencyPower Efficiency7001400210028003500SE +/- 26.53, N = 5SE +/- 26.57, N = 6SE +/- 27.73, N = 4SE +/- 18.15, N = 42641263626613311

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: Zxing 1D/2D Barcode Image ProcessingStockJava ThroughputJava LatencyPower Efficiency130260390520650SE +/- 4.02, N = 10SE +/- 3.60, N = 14SE +/- 3.12, N = 10SE +/- 4.50, N = 15558543544602

Sunflow Rendering System

This test runs benchmarks of the Sunflow Rendering System. The Sunflow Rendering System is an open-source render engine for photo-realistic image synthesis with a ray-tracing core. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSunflow Rendering System 0.07.2Global Illumination + Image SynthesisStockJava ThroughputJava LatencyPower Efficiency0.08330.16660.24990.33320.4165SE +/- 0.003, N = 15SE +/- 0.005, N = 15SE +/- 0.003, N = 15SE +/- 0.007, N = 150.2420.2590.3020.370MIN: 0.14 / MAX: 0.67MIN: 0.15 / MAX: 0.66MIN: 0.18 / MAX: 0.63MIN: 0.21 / MAX: 1.01