Innovation

A lot of technology leaders seem to think that a public cloud is unsafe for sensitive operations like banking and finance, which by extension would mean any industry involving commerce on the web, such as retail. You can engineer the Cloud Network the way you need with maximum compliance based on your regulatory needs.

by Author

Recently, the cloud based delivery strategy of IBS received a tremendous boost at the Amazon Web Services (AWS) Warriors event. As a pioneer in deploying mission critical applications on the AWS cloud, our digital travel solution stood out for the innovative architecture, which is cloud native in design. Honestly, a lot of the stakeholders, including prospective customers and current ones who could derive a tremendous amount of value from this, haven't fully grasped the magnitude of what this offers. Therefore let me explain in some detail the nuances of a cloud based software delivery model as it is today, as well as what it is evolving into in the near future.  

Glossary

IT provider: The vendor of a software product, which is delivered in SaaS mode for our context. This may be through a private cloud or a public cloud, both of which are explained below. Example: IBS Software

Cloud provider: The third party company which provides the infrastructure for a public cloud, and thus facilitates one particular type of SaaS mode delivery of the software product. These companies typically have very large amount of resources, most importantly server space capable of accommodating several hundred thousand times the maximum requirement of each IT provider. This is further explained below. Examples: Amazon Web Services, Microsoft Azure

Customer: The company, such as an airline or a travel company, which is a customer of the IT provider and makes use of the software product as a platform for its business. As is common in SaaS mode implementations, they are billed per transaction made by the end user (passenger or his/her travel agent) through the software platform. Example: Any airline in the world! 

Look to book ratios: Recalibrating performance expectations

When technology is evolving to a point where the global tally of mobile phone subscriptions outnumber the population of the world, this means that there is an unprecedented level of accessibility for the services which we provide through the internet – and travel reservations are no different. The number of people who are searching for travel products is significantly more than the number of actual bookings that will materialize from this. The difference lies in the "look to book" ratio, which has enormous impact on the technology resources demanded by a typical travel company's business system.

It is estimated that the this ratio ranges from 10,000:1 to 50000:1 in the modern world even this is growing fast. This essentially means that for every purchase that happens on a PSS (for which the technology provider receives a fee) there are at least 9,999 other transactions (merely user queries) which don't generate any revenue for either party because no purchase happens. So a customer expecting 10,000 bookings in a given time frame must actually prepare its systems to handle 10,000 times 10,000 which is 100 million transactions in the same time frame. Despite the tremendous drain on system resources, these empty transactions are a very real part of the sales process and must necessarily be handled by the system. 

What makes current SaaS based solutions less than perfect?

Many technology companies which offer cloud based software products to their customers do so through a private cloud, which is created using their fully owned data centres, or leased hardware which is totally under their control. At IBS, our range of products are delivered in SaaS mode, for which we have a total of many regional data centres around the world. These are maintained by us directly, and the specific data centre used to feed a customer's software implementation is decided based on geography. Customers are serviced based on the nearest geographic location for minimal latency and also data locality preferences. Customers are assured of their data being fully secured, since the cloud is fully behind our engineered networks and managed by our in-house team of experts. From a customer's perspective, this is ideal – they are billed on the basis of number of transactions, without any concerns regarding scalability.  

For instance, take an LCC named Example Airways, which serves a small market with a customer base of around 100,000 people. During peak time, which is an average of one hour per day, the system needs to handle as many as 25,000 queries, while at other times it is only in four figures or less. To the customer, the SaaS model offered by IBS is an ideal solution, because it does not need to set up its own IT infrastructure to manage those fluctuating volumes. Instead, the cloud is already provisioned with enough system resources to serve a peak load of 25,000 queries, although we are well aware that for 23 hours of every day these are lying idle. Next quarter, if Example Airways becomes so popular that their passenger base and peak load both become double of what it is today, what happens? The customer continues to pay our fee on a per transaction basis, maintaining their cost predictability, but we need to expand our own data centre infrastructure and deploy new servers to support this volume increase.

In other words, the "advantages" of the private cloud are passed on to the customer only because the inherent risks are absorbed by the cloud owner, which in this case is the IT provider. However, IT providers like IBS Software are in the business of software products, and not cloud management per se. So there are times when the scale demanded by the customer goes beyond the scope of a private cloud. This is where we introduce the public cloud.

Watch for a deeper dive into the public cloud and its advantages over the private cloud in Part 2 of this blog series