Comparison Of PaaS Cloud Service Providers – Google App Engine, AppFog, Red Hat OpenShift, And Heroku

Cloud Service Providers

The notion “cloud computing” invokes the access of data or other services from other scalable information centers over the network from any remote location (Gartner, 2014). The contemporary model of computing contains boundless potential within it that can be utilized for all the sectors. In cloud services, the users are given permission to access the databases and application software. The cloud service providing enterprises handle the architecture and cloud platforms on which the applications are executed. 

Based on the services provided by the cloud traders, the cloud computing services are categorized into Software-as-a-Service (SaaS), PaaS, Infrastructure as a Service (IaaS), etc.

The PaaS cloud service platforms permits the application developers to organize and control their developed web applications without the need for creating and managing the complicated infrastructure (Grossman, 2009).

The PaaS platform helps the web developers to focus only on the code and not the application management. So they can build the applications without having issues on IT functions or cannot have to wait for the server to get ready for writing the source code.

The top-most PaaS service providers are described below:

Google App Engine

This is a PaaS platform that allows the users to code and execute the applications in Google’s framework. The applications are easily drafted, maintained, and they are scalable since the modification in the storage and network traffic is required.

Main Features:

• A sample application can be created and executed in the cloud promptly
• The application runs in a secure sandboxed scenario that is not dependent on the OS, hardware components, or any server physical location

AppFog

It is a multi-architecture and multi-language PaaS platform used for framing several private clouds.

Main Features:

• It lacquers the cache and hence the applications can be executed with less server load and efficient performance
• It supports many web application technologies like PHP, Ruby, Java, Python, and Node (Sloan, 2009)
• The familiar services like MySQL, Redis, RabbitMQ, and PostgreSQL can be added to the user’s application

Red Hat OpenShift

This depends on the open source applications and provides many programming languages, modules, and databases. It can be customized by the users and are available in three forms:

1. A cloud-based Hosting Service called OpenShift Online.
2. A private PaaS platform running in the user’s data center called as OpenShift Enterprise.
3. The open-source application host platform is termed as OpenShift Origin.

Main Features:

• The time required for application deployment can be reduced by letting the users to concentrate on the code and creativity rather than framework management
• It provides wide variety of frameworks, runtimes, and programming languages inclusive of JBoss EAP and Java EE6 to the users

Heroku

This platform offers two computing environments namely worker dynos that is constructed for task request receipt and web dynos that reacts to HTTP requests. It also offers third party applications and is compliant to work with the applications formed by utilizing the Twelve Factor App Methodology.

Main Features:

• It permits you to administer the environment related arrangements like backing credentials individually form the user’s code so as to ensure safety.
• The Ad-hoc commands can be executed by single-off application environment prototype.
• By default, it supports programming and scripting languages like Ruby, Python, PHP, Node.js, and Java. A customized build pack can be used to build other applications also.

When compared with Bluemix, it saves time of the user by allowing them to focus on the application itself rather than the configuration, installation, and security of the application. The Heroku stands 3^rd position with 1,931 followers when compared to Bluemix with a rank 39^th in PaaS clouds.

Features	Heroku	Bluemix
Data Storage	It supports ClearDB MySQL, Redis, MangoDB, IronCache, Graphene DB, and Heroku Postgres	It supports Cloudant, ClearDB, MangoDB, Redis, mySQL, Elephant SQL,PostgreSQL for data storage
Integrated Runtime Options	Java, Ruby, Node.js, Clojure, Scala, Python	Java, Node.js, Noop, Ruby on Rails, Ruby Sinatra

 
Table 1: Comparison of Bluemix with Heroku

The two PaaS cloud platforms BlueMix and Heroku are compared based on different perspectives as summarized in the tables below.

Comparison of Bluemix and Heroku based on smart application development services

	Heroku	BlueMix
Integrated Services	It is mostly third-party and contains wide range of add-on services, search, communication, workload administration, logging, media, payment control, etc.	It supports partly third-party applications along with location and performance tools, Big Data administration, information security, and Internet of Things (Sloan, 2009)
Openness	They can run on Amazon Cloud but certain flexibility is imposed for Heroku users	It is an execution of the IBM’s open cloud framework based on the open source PaaS platform Cloud Foundry

 
Table 2: Comparison of Bluemix and Heroku based on smart application development services

Bluemix Vs Heroku based on quality of design and navigation of the PaaS System

	Heroku	BlueMix
Quality of Design	The teams are enabled for receiving more application updates for creating the applications quickly. The generated application quality can be increased by means of earlier bug fixing and reliable testing	It adopts supple methods for creating quality applications. The user can design the BlueMix solutions based on their own decisions on BlueMix services
Navigation	Deployment of the application can move the app from a local system to PaaS system	By just clicking on the “Deploy” option, the application gets deployed easily. This leads to a full featured and fully automated deployment solution while using PaaS

Table 3: Bluemix Vs Heroku based on quality of design and navigation of the PaaS System

The projects created by the users reside on the BlueMix DevOps Services.

Step 1: Creating project for the app

1. Log in to IBM BlueMix DevOps Services.
2. Click My Projectson the top and select CREATE PROJECT.
3. In the Create a projectpage, key in the project name, select Make this a Bluemix Projectbox and then click create.

Step 2: Creating simple webpage

1. Click EDIT CODE button in the project’s overview page.
2. Click File->New->File in the Web IDE.
3. Provide name for the highlighted file and enter index.html.
4. In the editor area on the right side of the pane, enter the following

<html>

     <body>

          Hello World!

     </body>

</html> 

Step 3: Create the manifest file by entering manifest.yml.

Step 4: Deploy and open the application

Comparison of BlueMix and Heroku based on the development and deployment complexities

	Heroku	BlueMix
Development	The Heroku applications are developed and administered with the help of client software called Heroku Toolbet that should be installed by the developer (Kaufman, 2009). The development complexity arises since additional software are required to be installed for every programming language	A microservice-based solution called “OpenWhisk” is introduced for reducing the complexity of the application development. It allows the application developers to form solutions using modern chaining and impose charges only the executed code
Deployment	The Version control system keeps track of many versions of code and hence the changes to the source code are coordinated among the web development teams of all sizes. Multiple deployment environments like staging, Quality Assurance testing, and production are maintained by formulating a Heroku app for each deployment environment thereby reducing the deployment complexity for larger applications	An excellent deployment group is provided with BlueMix and the solutions are released as an unit of the UrbanCode portfolio for reducing the complexity and to ensure the proper execution of the applications

 
Table 4: Comparison of BlueMix and Heroku based on the development and deployment complexities

Comparison of BlueMix and Heroku based on the Costing Model

	Heroku	BlueMix
Costing	It offers free charge for sandbox applications, which is idle for most of the times. For smaller applications, the price starts at $ 7 per dyno for one month and reaches till $ 25+ per dyno at complex levels (Knights, 2009). In addition, the user has to pay 20 dollars per month for the utilization of Secure Socket Layer (SSL) on the user’s website	After 30 days of free trial, the cost is imposed. For Enterprise base cluster, the defined cost is $ 9,800. For base cluster of standard capacity, the cost is $ 7,100 and includes 384 RAM memory, 1Gbps networking speed, and 96 cores

Table 5: Comparison of BlueMix and Heroku based on the Costing Model

Comparison of BlueMix and Heroku based on the Application Management

	Heroku	BlueMix
Resource Management	The dyno manager makes dynos to run automatically. Hence the app operation can be easily maintained and the resources are managed well.	It provides the dashboard on which the creation, management, and display of available services and applications are done by the application developers, while monitoring the utilization of the resources.
Data Center Management	The servers are scalable and the better performance are attained with less expertise since the servers are managed by the users itself.	Each data center of the Bluemix has experts for troubleshooting and solving only the rare issues that are not directly solved by the automated management system.

Table 6: Comparison of BlueMix and Heroku based on the Application Management

Based on the analysis, BlueMix and Heroku platforms are greater options for startup enterprises. Heroku contains many integrated add-on features and easier deployment. From the cost perspective, Heroku services are cheaper than Bluemix. Bluemix can be the better option since it comparatively provides more deployment services than Heroku. Moreover, Heroku cloud platform purchased by Salesforce in 2010 was mainly developed for Ruby based applications but after that it added support for Java, Python, Node, PHP, Clojure, and Scala.

Conclusion:

Bluemix mostly offers similar services as Heroku. The easier deployment and add-ons are an added advantage. The Bluemix platform can be selected for Watson Application Programming Interface (API) since it contains some inexpensive services ranging from speech, text, document conversion, document translation, etc. The additional benefit of Bluemix is that it has Cloud Foundry together with the Docker container support and OpenStack virtual systems (Hutchinson, Ward & Castilon, 2009). The total pricing is similar to Heroku but gets more expensive when you scale the application.

References:

Gartner, M. (2014). Platform as a Service: Definition, Taxonomy and Vendor Landscape. Retrieved 11 June 2011, from https://www.gartner.com/doc/2833022/platform-service-definition-taxonomy-vendor.

Grossman, R. (2009). The Case for Cloud Computing. IT Professional, 11(2), 23-27.

Hutchinson, C., Ward, J & Castilon, K. (2009). Navigating the Next-Generation Application Architecture. IT Professional, 11(2), 18-22.

Kaufman, L. (2009). Data Security in the World of Cloud Computing, IEEE Security and Privacy, 7(4), 61-64.

Knights, M.  (2009). Can security kill cloud computing?. Retrieved 10 June 2010, from https://www.itpro.co.uk/610299/can-security-concerns-killcloud-computing.

Sloan, K. (2009). Security in a virtualised world. Network Security. 2009(8), 15-18.