svn commit: r909390 - in /websites/staging/stratos/trunk/content: ./ about/why-apache-stratos.html

[bu...@apache.org]

Author: buildbot
Date: Tue May 20 09:50:19 2014
New Revision: 909390

Log:
Staging update by buildbot for stratos

Modified:
    websites/staging/stratos/trunk/content/   (props changed)
    websites/staging/stratos/trunk/content/about/why-apache-stratos.html

Propchange: websites/staging/stratos/trunk/content/
------------------------------------------------------------------------------
--- cms:source-revision (original)
+++ cms:source-revision Tue May 20 09:50:19 2014
@@ -1 +1 @@
-1596166
+1596169

Modified: websites/staging/stratos/trunk/content/about/why-apache-stratos.html
==============================================================================
--- websites/staging/stratos/trunk/content/about/why-apache-stratos.html (original)
+++ websites/staging/stratos/trunk/content/about/why-apache-stratos.html Tue May 20 09:50:19 2014
@@ -153,7 +153,7 @@
 <p>If you look at current PaaS offering, you can't get true flexibility for your own custom service. Many cases, you have to bare with what they are offering and make your solution around it.  With the Apache Stratos cartridge model, you can create custom service, without any limited boundaries. Your cartridge can be fully configured (installed all software, configuration ..etc) or zero configured (let cartridge user to install and config what they want) or in-between.  This will allow you to customise your PaaS to suite in your current business workflows.</p>
 <h2>Multi-factored auto scaling</h2>
 <p>Majority of PaaS providers only used http traffic to auto scale or dynamic resource provisioning. It gives limitation on scaling non-http based application. But Apache Stratos auto scaler can take http traffic and non-http health data to do intelligent dynamic resource provisioning. Also its capable of dealing with these multiple factors run through scaling algorithm simultaneously. Inflight request count on Load Balancers, Load average, Memory consumption are some of factors and its easy to extend (benefit from clean unified communication) to any factors need to take care of the auto scaling.</p>
-<h2>Predicting future loads</h2>
+<h3>Predicting future loads</h3>
 <p>Another important capability of Apache Stratos is predicting immediate future load, based on current health statistics. All cartridges, Load balances, health monitors are publishing health status into CEP (Complex Event Processor) via real time event bus. CEP will aggregate all these events and calculate first derivative, second derivative for a given time window frame and paas resulting streams to Auto scaler via reliable message bus.</p>
 <p>Auto Scaler used motion equation to get immediate (next minute) future load.</p>
 <span>s = ut+ 1/2 at<sup>2</sup></span>
@@ -177,9 +177,38 @@
 
 <p>Apache Stratos capable of provision scalable load balancers for any IaaS clouds well as integrate any third party load balancers. This gives more flexibility and extendability while providing more efficient load balancing based on deployment environment.</p>
 
-<h2>Apache Stratos Load Balancing features and benefits</h2>
+<h3>Apache Stratos Load Balancing features and benefits</h3>
 
 <p>Provisioning - Load balancers can spin up dynamically. Based on load balancer as cartridges, Apache Stratos will spin up defined minimum load balancers when the first application subscription comes. Also capable of defining load balancers to spin up in every cloud, regions enabling effective multi-cloud deployment specially for geo-graphical based application deployments.</p>
+
+<p>Flexibility - Can defined service level dedicate load balancing. Easy to define service level load balancers via REST API while capable of doing load balancing multiple services in single load balancer.</p>
+
+<p>Expandability - Capability of integrate any third party load balancers. With the message broker and topology based model, its easy to integrate load balancers like HAProxy, nginx, AWS ELB ..etc.  it gives optimize load balancing based on the deployment. (e.g use AWS ELB on EC2 deployment)</p>
+
+<h2>Capability of controlling IaaS resources</h2>
+
+<p>Different IaaS vendors provide different resource pooling, like regions, availability zones, hosts, subnets etc. If we could not controlled these resources and utilized in PaaS layer , leads to waist many resources. Also it is very important controlling them to do high available kind of deployments.</p>
+
+<p>Apache Stratos has carefully designed to address above problem. Apache Stratos has introduced partitions, network partitions to group IaaS resources to deploy cartridges in very controlled manner.</p>
+
+<h3>Partitions</h3>
+
+<p>Partitions are logical group, that can defined by devOps based on available IaaS resource pooling like regions, availability zones ..etc.  With this definitions, Apache Stratos capable of selecting where to spawn Cartridge instances, selected resource pool which is very useful in deployment like high available or disaster recovery.</p>
+
+<h3>Network Partitions</h3>
+
+<p>With this concept, Apache Stratos can logically group one or more partitions that are reside in same network.  With this network partition concept, Apache Stratos Auto Scaler can have intelligence of monitor and predict load of particular network bound regions and take necessary auto scaling decisions very easily. This also very unique and important feature Apache Stratos bringing into PaaS world.</p>
+
+<p>Smart Policies</p>
+
+<p>With the partitions and Network partitions, devOps can defined smart policies namely auto scaling and deployment policies, which can be used in cartridge subscriptions. Smart policies will provide cloud SLAs which absolutely mandatory when you consider cloud solutions.</p>
+
+<h2>Logging, metering and monitoring</h2>
+
+<p>Logging, metering and monitoring are primary but important features you should not neglect. Apache Stratos capable of capturing all health statistics, application logs well as usage and aggregate them into centralized location. it provide devOps life easy otherwise it is impossible task to login and monitor each and every instances in a large scale deployment which has hundreds of instances.</p>
+
+<h2>Cloud bursting</p>
+<p>Cloud bursting is now widely discuss since it reduces TCO (Total Cost of Ownership) while provisioning application into another cloud to handle peak load. Like all other aspect, Apache Stratos cloud bursting also carefully design to work with private, public and hybrid cloud with effective provisioning load balancers per bursting cloud.</p>
         	</article>
     	</section>
   	</div>