<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic SGX benchmark test in Intel® Software Guard Extensions (Intel® SGX)</title>
    <link>https://community.intel.com/t5/Intel-Software-Guard-Extensions/SGX-benchmark-test/m-p/1130949#M1893</link>
    <description>&lt;P&gt;I am doing a benchmark of ecall.But I came across a problem.&lt;/P&gt;

&lt;P&gt;Due to the size limit of enclave, there will be a &lt;STRONG&gt;page thrashing&amp;nbsp;&lt;/STRONG&gt;when the data passed into the enclave exceeds the size of enclave.In&amp;nbsp; my&amp;nbsp;&lt;SPAN style="font-size: 1em;"&gt;experiment,&lt;/SPAN&gt;&lt;SPAN style="font-size: 1em;"&gt;different size of array is passed into enclave via ecall to be sorted using quicksort. I record the time the ecall used.&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;Below is opaque's(&lt;A href="https://people.eecs.berkeley.edu/~wzheng/opaque.pdf"&gt;https://people.eecs.berkeley.edu/~wzheng/opaque.pdf&lt;/A&gt;)&amp;nbsp;sort benchmarks. It can be seen that when the data is larger than the EPC,there will be a "jump".&lt;/P&gt;

&lt;P&gt;&lt;span class="lia-inline-image-display-wrapper" image-alt="opaque.png"&gt;&lt;img src="https://community.intel.com/t5/image/serverpage/image-id/9951iD8E99C320E3A98F0/image-size/large?v=v2&amp;amp;px=999&amp;amp;whitelist-exif-data=Orientation%2CResolution%2COriginalDefaultFinalSize%2CCopyright" role="button" title="opaque.png" alt="opaque.png" /&gt;&lt;/span&gt;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;Sadly,below is my benchmark.It is linear:&lt;/P&gt;

&lt;P&gt;&lt;span class="lia-inline-image-display-wrapper" image-alt="own.PNG"&gt;&lt;img src="https://community.intel.com/t5/image/serverpage/image-id/9952iD02E61FA3E189082/image-size/large?v=v2&amp;amp;px=999&amp;amp;whitelist-exif-data=Orientation%2CResolution%2COriginalDefaultFinalSize%2CCopyright" role="button" title="own.PNG" alt="own.PNG" /&gt;&lt;/span&gt;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;How does it happen?&amp;nbsp; Thanks!&lt;/P&gt;

&lt;UL&gt;
	&lt;LI&gt;CPU is&amp;nbsp;Intel(R) Core(TM) i5-7400, 8 GB RAM.&lt;/LI&gt;
	&lt;LI&gt;SGX sdk and psw are both 2.2.&lt;/LI&gt;
	&lt;LI&gt;Experiment is under HW mode.&lt;/LI&gt;
	&lt;LI&gt;The enclave max heap size is 16MB.&lt;/LI&gt;
&lt;/UL&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Tue, 18 Sep 2018 09:01:36 GMT</pubDate>
    <dc:creator>zhou__chen</dc:creator>
    <dc:date>2018-09-18T09:01:36Z</dc:date>
    <item>
      <title>SGX benchmark test</title>
      <link>https://community.intel.com/t5/Intel-Software-Guard-Extensions/SGX-benchmark-test/m-p/1130949#M1893</link>
      <description>&lt;P&gt;I am doing a benchmark of ecall.But I came across a problem.&lt;/P&gt;

&lt;P&gt;Due to the size limit of enclave, there will be a &lt;STRONG&gt;page thrashing&amp;nbsp;&lt;/STRONG&gt;when the data passed into the enclave exceeds the size of enclave.In&amp;nbsp; my&amp;nbsp;&lt;SPAN style="font-size: 1em;"&gt;experiment,&lt;/SPAN&gt;&lt;SPAN style="font-size: 1em;"&gt;different size of array is passed into enclave via ecall to be sorted using quicksort. I record the time the ecall used.&lt;/SPAN&gt;&lt;/P&gt;

&lt;P&gt;Below is opaque's(&lt;A href="https://people.eecs.berkeley.edu/~wzheng/opaque.pdf"&gt;https://people.eecs.berkeley.edu/~wzheng/opaque.pdf&lt;/A&gt;)&amp;nbsp;sort benchmarks. It can be seen that when the data is larger than the EPC,there will be a "jump".&lt;/P&gt;

&lt;P&gt;&lt;span class="lia-inline-image-display-wrapper" image-alt="opaque.png"&gt;&lt;img src="https://community.intel.com/t5/image/serverpage/image-id/9951iD8E99C320E3A98F0/image-size/large?v=v2&amp;amp;px=999&amp;amp;whitelist-exif-data=Orientation%2CResolution%2COriginalDefaultFinalSize%2CCopyright" role="button" title="opaque.png" alt="opaque.png" /&gt;&lt;/span&gt;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;Sadly,below is my benchmark.It is linear:&lt;/P&gt;

&lt;P&gt;&lt;span class="lia-inline-image-display-wrapper" image-alt="own.PNG"&gt;&lt;img src="https://community.intel.com/t5/image/serverpage/image-id/9952iD02E61FA3E189082/image-size/large?v=v2&amp;amp;px=999&amp;amp;whitelist-exif-data=Orientation%2CResolution%2COriginalDefaultFinalSize%2CCopyright" role="button" title="own.PNG" alt="own.PNG" /&gt;&lt;/span&gt;&lt;/P&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;

&lt;P&gt;How does it happen?&amp;nbsp; Thanks!&lt;/P&gt;

&lt;UL&gt;
	&lt;LI&gt;CPU is&amp;nbsp;Intel(R) Core(TM) i5-7400, 8 GB RAM.&lt;/LI&gt;
	&lt;LI&gt;SGX sdk and psw are both 2.2.&lt;/LI&gt;
	&lt;LI&gt;Experiment is under HW mode.&lt;/LI&gt;
	&lt;LI&gt;The enclave max heap size is 16MB.&lt;/LI&gt;
&lt;/UL&gt;

&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Tue, 18 Sep 2018 09:01:36 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Software-Guard-Extensions/SGX-benchmark-test/m-p/1130949#M1893</guid>
      <dc:creator>zhou__chen</dc:creator>
      <dc:date>2018-09-18T09:01:36Z</dc:date>
    </item>
    <item>
      <title>Hi, </title>
      <link>https://community.intel.com/t5/Intel-Software-Guard-Extensions/SGX-benchmark-test/m-p/1130950#M1894</link>
      <description>&lt;P&gt;Hi,&amp;nbsp;&lt;/P&gt;&lt;P&gt;I am doing the same benchmark as yours, but when I passed data which exceeds the enclave size, I get&amp;nbsp;Aborted (core dumped). Is there any method to avoid this issue?&lt;/P&gt;&lt;P&gt;Besides, are you find out the reason of your benchmark data (linearly)?&lt;/P&gt;&lt;P&gt;Thank.&lt;/P&gt;</description>
      <pubDate>Wed, 29 May 2019 02:16:19 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Software-Guard-Extensions/SGX-benchmark-test/m-p/1130950#M1894</guid>
      <dc:creator>lim__wen_sheng</dc:creator>
      <dc:date>2019-05-29T02:16:19Z</dc:date>
    </item>
  </channel>
</rss>

