Breadcrumb Breadcrumb

Web Content Display Web Content Display

 

Asset Publisher Asset Publisher

Additional Career Openings

07 Apr 2016 Openings for Software Engineer Somerset, NJ,07 Apr 2016 Sofia Technology, LLC has openings for Software Engineers in Somerset, NJ and other...

Spark Dataframes – From machine readable to human fathomable code

Background Apache Spark , an in-memory cluster compute platform that provides over 100x performance for certain applications over traditional MapReduce, has hit home with data analysts and...

Careers-Content-Testimonial

  Why We need you? We have an exceptionally great team at SOFIA and you now have an opportunity to play a key role as well. You will be joining a team in which each and every...

Shailesh Dangi

Shailesh Dangi Managing Partner & CTO Lorem Ipsum

Portal Technology Areas

Oracle Web Center Liferay IBM WebSphere Portal

SOFIA Liferay Competencies

SOFIA Liferay Competencies SOFIA has successfully led multiple portal projects by providing the required " thought leadership " that goes beyond product customization, delivering industry...

401(k) Qualified Retirement Plan

401(k) Qualified Retirement Plan SOFIA provides eligible employees with a 401(k) Qualified Retirement plan which is an excellent means of long-term...

Career-Grid-Blocks

we're glad you are here Armed with inveterate convictions to succeed, we at SOFIA share an entrepreneurial culture, where employees think and act like owners. Our...

Big-Data-Page-Banner

Ignite your Big Data Projects with a "Spark" Real-Time Predictive Analytics with Spark Streaming

Liferay Platinum Partner

Liferay Platinum Partner SOFIA's partnership with Liferay provides its clients a foundation for web delivery that has one of the lowest TCO with a quicker time-to-market...

Liferay Recognizes SOFIA Technology as a Platinum Partner

22 Oct 2013 Liferay Recognizes SOFIA Technology as a Platinum Partner Partnership continues to grow between Liferay and enterprise portal services leader in financial...

SOFIA Liferay Quote

Over the past three years, the SOFIA team has been a leader in bringing Liferay implementations to the financial services space. We look forward to our strengthened partnership with them.

SOFIA Liferay Services

Liferay Liferay, a Leader in Gartner’s Magic Quadrant for Horizontal Portals, is one of the the fastest-growing Java-based portal, increasing its market momentum even as other Java portals...

Case Study

Client Success & Case Study At SOFIA, we like to share some of the key constituents for client success – combining our technical depths along with...

Asset Publisher Asset Publisher

Back

Custom Caching Framework based on Portlet Filters

 

SOFTWARE:

 

  1. ehCache
    (Download: http://sourceforge.net/projects/ehcache/files/ehcache-core )

  2. Liferay 6.0.3
    (Download:
    http://sourceforge.net/projects/lportal/files/Liferay%20Portal/6.0.4/liferay-portal-tomcat- 6.0.4.zip/download)

  3. Filter Source code (Download: )

BACKGROUND:

Introduced as part of JSR-286, portlet filters dynamically intercept requests from and responses back to the portlet container. Just like Servlet filters, portlet filters can be leveraged for caching responses to improve the performance.

The following article presents how combining portlet filters and Terracotta’s in-memory ehCahe can boost the portlet performance, in turn reducing the back-end traffic and providing blazing performance to deliver content. ehCache provides convenient REST based APIs to populate and invalidate cache that applications surfaced as portlets could leverage.

Ehcache:

Ehcache is an open source, standards-based cache used to boost performance, offload the database and simplify scalability. Ehcache is robust, proven and full-featured and this has made it the most widely- used Java-based cache.

It can scale from in-process with one or more nodes through to a mixed in-process/out-of-process configuration with terabyte-sized caches. For applications needing a coherent distributed cache, Ehcache uses the open source Terracotta Sever Array.

Ehcache is actively developed, maintained and supported as a professional open source project by Terracotta, Inc. and is available under an Apache 2 license.

Request/Response flow:

  1. End-User requests for a page containing portlets

  2. Cache Filter intercepts the request, initializes the filter on the first request, extracts and

    instantiates the necessary classes based on the initialization parameters defined in portlet.xml

    file

  3. Filter checks the cache against a user defined cache key implementation and for a cache-miss

    sends the request to the portlet along with a wrapped response, retrieves the mark-up, and,

  4. Caches the content before sending the response to the next filter in-line on the way back to the

    portlet container

    All subsequent requests from the portlet container to the portlet are served from the cache, until an external event happens that invalidates the cache or if cache expires.

ehCache Installation:

http://ehcache.org/documentation/documentation-1_1.html#mozTocId747622

ehCache Configuration file:

<?xml version="1.0" encoding="UTF-8"?>
<ehcache xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"

xsi:noNamespaceSchemaLocation="ehcache.xsd" updateCheck="true" monitoring="autodetect" dynamicConfig="true" >

<diskStore path="java.io.tmpdir"/> <defaultCache

maxElementsInMemory="0" eternal="false" timeToIdleSeconds="1200" timeToLiveSeconds="1200">

    </defaultCache>

<!--

This cache contains a maximum in memory of 10000 elements, and will expire an element if it is idle for more than 5 minutes and lives for more than 10 minutes.

If there are more than 10000 elements it will overflow to the
disk cache, which in this configuration will go to wherever java.io.tmp is defined on your system. On a standard Linux system this will be /tmp"
-->

<cache name="Acme_WDFCache" maxElementsInMemory="10000"

maxElementsOnDisk="1000" eternal="false" overflowToDisk="true" diskSpoolBufferSizeMB="20" timeToIdleSeconds="300" timeToLiveSeconds="600" memoryStoreEvictionPolicy="LFU"

/>
</ehcache>

Filter Classes:

[jboss@daffy /opt/app/liferay/liferay-portal-5.2-ee-sp3/tomcat- 6.0.18/webapps/WDF_UCM/WEB-INF/classes/com/schoolspecialty/wdf/filter/portlet]$ls -ltr total 20

-rw-rw-r-- 1 jboss jboss 936 Jul 21 16:22 CachedContent.class
-rw-rw-r-- 1 jboss jboss 2353 Jul 21 16:22 CachedRenderResponseWrapper.class -rw-rw-r-- 1 jboss jboss 7559 Aug 6 14:25 PortletCacheFilter.class -rw-rw-r-- 1 jboss jboss 2578 Aug 6 16:14 CachedOutputStream.class

[jboss@daffy /opt/app/liferay/liferay-portal-5.2-ee-sp3/tomcat- 6.0.18/webapps/WDF_UCM/WEB-INF/classes/com/schoolspecialty/wdf/cache]$ls -ltr total 16
-rw-rw-r-- 1 jboss jboss 1872 Jul 21 16:20 SSIPortletsCacheKey.class -rw-rw-r-- 1 jboss jboss 2474 Jul 21 16:20 EHCacheUtil.class
-rw-rw-r-- 1 jboss jboss 493 Jul 21 16:20 CacheUtil.class
-rw-rw-r-- 1 jboss jboss 372 Jul 21 16:20 CacheKeyIntf.class

CacheKeyIntf<T> defines a single method getKey that returns a client implemented key of type T Portlet Mapping File:

<filter>

<filter-name>CacheFilter</filter-name> <filter-class>com.acme.filter.portlet.PortletCacheFilter</filter-class> <lifecycle>RENDER_PHASE</lifecycle>

<init-param> <name>CACHE_ALGORITHM</name> <value> x.x.x.EHCacheUtil</value>

</init-param>
<init-param>

<name>CACHE_NAME</name>

<value>Acme_WDFCache</value> </init-param>

<init-param> <name>CACHE_KEY</name>

<value>x.x.x.SSIPortletsCacheKey </value> </init-param>

</filter> <filter-mapping>

<filter-name>CacheFilter</filter-name>

<portlet-name>1</portlet-name> </filter-mapping>

Response Times:

1) Response is actually created for a UCM Portlet and returned (not cached): 22 ms avg 2) Response is returned from cache: 2 ms avg