Why Software as a Service Is Going to Dominate the Next Several Years - Case Study Example

Why Software as a Service Is Going to Dominate the Next Several Years Abstract Cloud Computing is being talked today almost everywhere across the globe in the IT industry. It is revolutionising the way we think of computing. Software as a Service or SaaS is making inroads in vital areas especially in enterprise organisations. In this paper, we are going to discuss how SaaS is going to dominate the next several years in the field of Information Management. Cloud computing is a general term used by IT professionals for the delivery of computing requirements as services over the Internet. Clouds can be very similar to a rental car agency, in that you can rent cloud service for a short period of time. These services can be broadly classified into three categories: Software as a Service (SaaS), Platform as a Service (PaaS) and Infrastructure as a Service (IaaS). SaaS is by far the most widely used service of the cloud in today’s time and hence, cloud computing and SaaS are used synonymously sometimes even though SaaS is actually one of the services provided by cloud. Considering the current trends, SaaS is most likely to dominate the next several years in the field of information management. SaaS is a software delivery model often referred to as “on demand software” in which the software and the required data are centrally hosted on the cloud and are typically accessed by the users over the internet using a thin-client such as a web browser. One daily life example of this is the famous Google Docs in which one is able to open Microsoft Word document files, PDF files within the browser itself without having to download the software to your computer and open the file using the installed application. The cloud computing concept is not actually new. Computing as a utility was developed as a part of MULTICS (Multiplexed Information and Computing Service) which was a time-sharing operating system developed way back in the 1960s. However, what is new in cloud computing is the illusion of infinite resources on demand, fine grained billing, and the concept of public and private clouds. Also the pay-as-you-go paradigm is what makes cloud different from the former ideas similar to cloud. Cloud computing democratises access to supercomputer capability. In the past, when researchers and scientists needed to perform extremely complex and demanding calculations, they needed supercomputers that had tremendous processing speed and very large memory. This would require a lot of monetary resources and the permission to use such supercomputers was only given to credible individuals. However, with cloud coming in, a student or academia can use its capabilities and a large amount of resources to perform his research or experiments without the need of possessing a high-cost, high-maintenance supercomputer. Also large software like Matlab or MapReduce can be used by students or other professionals without having the software to be purchased and installed on their PCs. Using SaaS, you just need one installation on a centralised cloud and then several instances of this installation can be used by multiple users. This is a huge saving particularly in cases where we need to install software on several computers. Also it puts students and academia on a more level playing field to have a high impact on industry. Hence the next Google, eBay or Amazon can come from a small team of entrepreneurs even without heavy software and hardware. This way cloud computing has progressed from 2008 to 2011 helping Web start ups, computer science researchers, enterprise applications, academic, scientific and high performance computing. Cloud computing is primarily “on demand” which means that you can add or remove resources as and when you need them. Two examples of where this facility was needed can be discussed. In 2001, a famous incident known as the CNN home page meltdown occurred on September 11 when the traffic to the website increased by 10 times within a span of 15 minutes due to which the website had to go offline. In another incident, in 2008, when Animoto was released as a Facebook application, its users doubled every 12 hours for 3 days. To cope with this demand, Animoto had to increase its servers from 50 to 3500 servers. In the presence of cloud, these situations could have been easily handled because of the scalability that cloud provides. The pricing of these resources is also very flexible and is highly advantageous. In cloud, if you want the processing power of 1000 CPUs for one hour or the processing power of 1 CPU for 1000 hours, you need to pay the same amount of money. However, in the absence of cloud and the services that it provides, you would need to purchase 1000 CPUs for a particular period of time which would be expensive and these CPUs would be of no use after the experiment or research. Hence, SaaS provides a more flexible and economic solution. However, considering everything mentioned above, a few questions can be raised against SaaS. The first question that can be raised is that web browsers are not interactive enough to support SaaS. This was true when web applications required a full page refresh in order to update its content. But with the advent of Javascript, AJAX and HTML5, these arguments can be put to rest. The second is integration of SaaS in current systems. Certain applications have been designed to work in a static environment where one knows the amount of memory available and the amount of memory required. These applications also need stability and reliability which cannot be offered on the cloud because you may not know which part of your code is running on which machine. Such applications cannot be transferred to a cloudy environment which is dynamically changing. Another issue that can be raised is customisation. Many SaaS applications are fairly “packaged”. However, customisation tools are maturing today which allows you to develop your own environment. Big companies would want to own their own software rather than rent it. This may be true but the flexibility is lost because of the huge amount of investment that would be put in order to purchase and install the software. Thus, in the future due to rapidly changing technologies, if the firm decides to switch technologies, this would become expensive. However in a cloud environment, no uninstallation is necessary. All you need is a credit card. The biggest issue however, that still needs to be answered convincingly by many cloud experts themselves is the issue of legacy and security. The issue can be dealt with by saying that IT security is all about trust. We need to trust our hardware and software manufacturers, operating system and software vendors and also our ISP. Any one of these can undermine your security and lead to loss and corruption of data and crashing of systems. However, SaaS takes it a step further. In the previous case, you can install firewalls or IDSs and build a resilient system that works even in the presence of such insecurity. But in this case, you need to trust your outsourcer completely. Once the application is deployed on the cloud, the control over it is reduced. In case one of the machines fails, the performance and results can have a negative impact. You do not want your data to be on some cloud computer that abruptly disappears. You do not want the company you are using to be sold to your direct competitor. You not only have to trust the outsourcer’s security, but also its reliability, its availability and its business opportunity. Considerable amount of research is done on how security can be increased in cloud and this situation is certainly improving. Many software giants are making credible SaaS offerings. Microsoft has released Microsoft Dynamics Online which is the first dynamics family application to be offered in a SaaS model. Oracle continues to grow its Oracle OnDemand solution set. Also, SAP’s Business ByDesign appears to be pure SaaS, on-demand offering that is operating independently from the SAP mothership. The involvement of these IT giants in the SaaS model is proof of its potential and its ability to do well in the future. There are also a few examples of clouds being used in heavy processing. The New York Times wanted to make its historical archives available for online access which needed the conversion of about 11 million articles into PDF files. Initial estimates outlined that hundreds of servers and about 4 Tb of storage would be necessary. But the IT team at the Times suggested Amazon Web Services and kicked off 100 EC2 instances and 4 Tb of S3 (Simple Storage Service) storage and the job was completed the next day for just about $240. Another example comes from Washington Post where Peter Harkins, a Senior Engineer at the Washington Post, used the Amazon Elastic Compute Cloud (Amazon EC2) to launch 200 server instances to process 17,481 pages of non-searchable PDF images into a searchable online library. The job was completed in nine hours for a final expense of $144.62 which shows how extremely cost-efficient and brilliant cloud computing is. Amazon Machine Images (AMIs) are preconfigured with an ever growing list of operating systems. The operating systems currently available to use with your Amazon EC2 are Red Hat Enterprise Linux, Windows Server 2008, Fedora Linux, OpenSolaris, Windows Server 2003 Oracle and many others. A logical question that comes to one’s mind that if cloud is so useful and has its roots back in the 1960s, then why wasn’t it previously thought of? The answer is simple. The trigger to the thinking about the cloud was the Web Space race. Many websites like Google, Yahoo and Amazon became extremely popular and they needed to expand their data centers. More users were driven by their growth. The economy of scaling, which means to buy more hardware at cheaper rates, also influenced the cloud style thinking. Also, the advent of more pervasive broadband with larger bandwidths and the emergence of the open source software community encouraged the SaaS model considerably. Modern day challenges such as parallelism in programming and big data applications like DNA sequencing can be solved using cloud. These applications deal with a huge amount of data that is generated everyday and hence scalability becomes an issue. However, cloud is always ready to handle such circumstances. Thus, weighing the benefits of Cloud Computing and SaaS in particular, it is pretty obvious that Cloud Computing is not just hype. It is most probably the next big thing in the IT industry. The time may not be too far when everything would be run just “in the cloud”. References Don Fornes. (2010, April 7). The Software as a Service Dilemma. Retrieved from http://blog.softwareadvice.com/articles/uncategorized/the-software-as-a-service-dilemma-104071/ Bruce Schneier. (2009, June 4). Cloud Computing. Retrieved from http://www.schneier.com/blog/archives/2009/06/cloud_computing.html Sinclair Schuller. (2011, January 10). 2011 Predictions for Cloud, SaaS, Multi-Tenancy and More!. Retrived from http://www.saasblogs.com/business/2011-predictions-for-cloud-saas-multi-tenancy-and-more/ Berkeley, University of California. (Producer). (2010). Computer Science 10 – Lecture 20: Cloud Computing. Retrieved from http://www.youtube.com/watch?v=46dHNduzkGY Berkeley, University of California. (Producer). (2010). Computer Science 10 – Lecture 20: The Potential of Cloud Computing. Retrieved from http://www.youtube.com/watch?v=mYhToBvKxEY&feature=relmfu Read More