oOKevinOo opened a new pull request, #2088:
URL: https://github.com/apache/incubator-celeborn/pull/2088
<!--
Thanks for sending a pull request! Here are some tips for you:
- Make sure the PR title start w/ a JIRA ticket, e.g. '[CELEBORN-XXXX]
Your PR title ...'.
- Be sure to keep the PR description updated to reflect all changes.
- Please write your PR title to summarize what this PR proposes.
- If possible, provide a concise example to reproduce the issue for a
faster review.
-->
### What changes were proposed in this pull request?
Support for small file caching to improve shuffle performance,The
architecture is as follows:
### Why are the changes needed?
Support for small file caching to improve shuffle performance
### Does this PR introduce _any_ user-facing change?
None
### How was this patch tested?
The test resource used is as follows:
<html xmlns:v="urn:schemas-microsoft-com:vml"
xmlns:o="urn:schemas-microsoft-com:office:office"
xmlns:x="urn:schemas-microsoft-com:office:excel"
xmlns="http://www.w3.org/TR/REC-html40";>
<head>
<meta name=ProgId content=Excel.Sheet>
<meta name=Generator content="Microsoft Excel 15">
<link id=Main-File rel=Main-File
href="file:////Users/11136313/Library/Group%20Containers/UBF8T346G9.Office/TemporaryItems/msohtmlclip/clip.htm">
<link rel=File-List
href="file:////Users/11136313/Library/Group%20Containers/UBF8T346G9.Office/TemporaryItems/msohtmlclip/clip_filelist.xml">
<style>
<!--table
{mso-displayed-decimal-separator:"\.";
mso-displayed-thousand-separator:"\,";}
@page
{margin:.75in .7in .75in .7in;
mso-header-margin:.3in;
mso-footer-margin:.3in;}
.font5
{color:windowtext;
font-size:9.0pt;
font-weight:400;
font-style:normal;
text-decoration:none;
font-family:等线;
mso-generic-font-family:auto;
mso-font-charset:134;}
tr
{mso-height-source:auto;
mso-ruby-visibility:none;}
col
{mso-width-source:auto;
mso-ruby-visibility:none;}
br
{mso-data-placement:same-cell;}
td
{padding-top:1px;
padding-right:1px;
padding-left:1px;
mso-ignore:padding;
color:black;
font-size:12.0pt;
font-weight:400;
font-style:normal;
text-decoration:none;
font-family:等线;
mso-generic-font-family:auto;
mso-font-charset:134;
mso-number-format:General;
text-align:general;
vertical-align:middle;
border:none;
mso-background-source:auto;
mso-pattern:auto;
mso-protection:locked visible;
white-space:nowrap;
mso-rotate:0;}
ruby
{ruby-align:left;}
rt
{color:windowtext;
font-size:9.0pt;
font-weight:400;
font-style:normal;
text-decoration:none;
font-family:等线;
mso-generic-font-family:auto;
mso-font-charset:134;
mso-char-type:none;
display:none;}
-->
</style>
</head>
<body link="#0563C1" vlink="#954F72">
host memory | host cpu | disk | service heap memory | service off-heap
memory | worker | master | spark total resource
-- | -- | -- | -- | -- | -- | -- | --
128G | 64 core | 11 * 7.1T | 12g | 40g | 5 | 3 | 25T/6400C
</body>
</html>
When turning off memory cache, test the performance as follows：
Duration of application execution：
The performance when I use memory for caching is as follows：
Duration of application execution：
The peak flow rate exceeds 20G, and the maximum time consumption is reduced
to less than 2min,up to 4x performance improvement.
I think there are two aspects to improving performance:
1. Using memory reduces requests to obtain local disk data
2. Reduced file open consume and improved disk sequential read capability
--
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]
