This project is benchmark for phobos XML library using JMH.
For launching benchmarks type sbt jmh:run -prof gc -rf csv in terminal in root
directory of the project. It will print results to terminal and save
them in jmh-result.csv file. Execution of this command takes plenty of time.
Benchmarking is comparing phobos with scalaxb.
PurchaseOrderBenchmark is benchmark measuring performance of decoding PurchaseOrder
XML document. Schema of this document is placed in the /src/main/xsd/po/po.xsd file.
You can see example of such document below.
<?xml version='1.0' encoding='UTF-8'?>
<purchaseOrder orderDate="1999-10-20">
<shipTo country="US">
<name>Alice Smith</name>
<street>123 Marple Street</street>
<city>Mill Valley</city>
<state>CA</state>
<zip>90952</zip>
</shipTo>
<billTo country="US">
<name>Robert Smith</name>
<street>8 Oak Avenue</street>
<city>Old Town</city>
<state>PA</state>
<zip>95819</zip>
</billTo>
<comment>Hurry, my lawn is going wild!</comment>
<items>
<item partNum="QEKkaHedxFt8UPdis">
<productName>9DprI</productName>
<quantity>-235701016</quantity>
<USPrice>0.529541179615654</USPrice>
<comment></comment>
<shipDate>InwnoUVd5MulLm8Amez</shipDate>
</item>
<item partNum="d32JMv3hF9">
<productName>WKEcI</productName>
<quantity>2094590288</quantity>
<USPrice>0.9758509262748334</USPrice>
<comment>qhGeEJad</comment>
<shipDate>ij</shipDate>
</item>
</items>
</purchaseOrder>Benchmark consists of several decoding runs with different sizes of items list.
Single benchmark name looks like benchmarking.PurchaseOrderBenchmark.{L}{N},
where {L} is library name (either phobos or scalaxb) and {N} is size of items list.
Each item adds 200B in average to documents.
Properties measured are
time per operation (benchmarking.SimpleTest.{L}{N}) in operation per second and
allocation rate (benchmarking.SimpleTest.{L}{N}:·gc.alloc.rate.norm) in
bytes per operation.
Results of benchmarking are saved in results directory.
