prosgarz35 commented on PR #3228: URL: https://github.com/apache/james-project/pull/3228#issuecomment-5945307552
> Hmm why not? I am interested in how far can this apply to our testing suite, and the actual build time gain. > > Cool idea to me, BTW. yeah i also think this can help with pretty long build (i see 5 hours) in some tests there boosts 100-300 times like this ## Benchmark Results Table | Metric | Physical Disk (Local FS) | Google Jimfs (In-Memory) | Performance Advantage | | :--- | :--- | :--- | :--- | | **Total Execution Time** | **14,849 ms** (~14.85 s) | **79 ms** (~0.08 s) | **188.0x faster** (99.5% reduction) | | **Throughput (Read + Write)** | **16.84 MB/s** | **3,164.56 MB/s** (~3.16 GB/s) | **188.0x higher throughput** | | **Cycle Rate (create-write-read-delete)** | **134.7 ops/s** | **25,316.5 ops/s** | **188.0x more operations/s** | | **Average Latency** | **7,373.58 µs** (~7.37 ms) | **29.67 µs** (~0.03 ms) | **248.5x lower latency** | | **Median Latency (p50)** | **6,995.40 µs** (~7.00 ms) | **23.00 µs** (~0.02 ms) | **304.1x lower latency** | | **95th Percentile Latency (p95)** | **9,852.20 µs** (~9.85 ms) | **46.60 µs** (~0.05 ms) | **211.4x lower latency** | | **99th Tail Latency (p99)** | **16,124.70 µs** (~16.12 ms) | **82.10 µs** (~0.08 ms) | **196.4x lower latency** | | **Worst Single Spike (Max Latency)** | **54,147.90 µs** (~54.15 ms) | **4,772.40 µs** (~4.77 ms) | **11.3x lower jitter** | -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
