Share this post:
In recent months, some prominent organizations have begun moving away from their traditional pattern of purchasing colossal, general-purpose infrastructure technology, and have begun looking towards personalized technology options. This well-publicized trend has also highlighted a wider change in the market, namely, the increasing importance of picking not only the right application, but also the right supporting technology to achieve the end goal.
The traditional database is far from ideal
At the end of the day, the costs of utilizing one-size-fits-all, closed-source databases can amount to more than just their sticker price. They require employment of dedicated server administrators to manage and maintain the databases and related resources, as well as acquiring the extra hardware to house the database software. Also, routine maintenance, updates and upgrades create additional complexity and potential complications as well as reduced availability of the database.
These generic, closed-source databases, can also be very inefficient compared to purpose-built data storage solutions for specific applications since, by their very nature they are designed to cover a wide variety of applications. To minimize these database’s deficiencies and limit system inter-dependencies that could impact their solutions running on top of these databases, Independent Software Vendor (ISV) may even be inclined to over engineer the recommended sizing, version and maintenance requirements to compensate for the lack of specialized performance design, adding yet another onslaught of high costs to customers.
End of the one-size-fits-all colossal database
Cost is not the only factor. The shift away from the one-size-fits-all database is being driven by the needs of those creating the applications as well. Rather than design their applications to the generic technologies, developers are selecting the individual ensemble of technologies to satisfy the application’s needs, including speed, scale and reliability. This is magnified by the necessity of a tailored approach while implementing payment fraud prevention as well.
The purpose-built real-time ‘dual-access’ datastore for payment fraud prevention
In the past, fraud patterns generally did not change much over time. Fraud prevention systems needed updated detection models only every few months. Downtime of a fraud system due to reconfiguration of the database, importing of a new fraud model, and so forth was tolerable since it was so infrequent and minimal.
No longer. Fraudsters are now continuously developing new fraud schemes and new variations on old themes. Once a working scheme is devised, fraudsters can roll it out very quickly on a large scale. To immunize against such perils, fraud prevention models not only must be developed in radically shorter time periods, but must also be deployed without any interruption of service. Never has agility and speed in fraud detection and prevention been more crucial. Therefore, to be effective, fraud prevention solutions need to perform in real time. That’s why IBM Safer Payments created a purpose-built NoSQL big-data in-memory datastore which is designed to be orders of magnitude faster than any generic database technology.
But, a real-time datastore, while an essential ingredient, may not be sufficient. Persistent bottlenecks may arise if vital access to it by the solution’s modeling and analysis functions are not sufficiently regulated and rationed. To achieve maximum optimization, IBM Safer Payments also introduced a fully-integrated ‘dual-access’ real-time database engine. This engine supports transaction processing concurrently with all modeling and analytics functions, such as model development, analysis, simulations. Thus, system functions operate resource-independently, with no function seeking additional resources nor is blocked by other resources, eliminating potential system bottlenecks.
IBM Safer Payments leads to increases in accuracy, speed
This massive performance optimization is the key behind IBM Safer Payments’ ultrafast statistical analysis and interactive reporting capabilities. We have heard from clients like Australian payments provider Indue that analyses that previously took hours on traditional platforms have increased in speed by 90%. (Don’t just take it from me. Check out their story here.) Because this ultra-fast database technology also propels the solution’s analytical capabilities, the performance of different decision models can be quickly compared and fraud countermeasures can be efficiently developed.
In some current IBM Safer Payments installations, fraud models are revised several times a day or more. This is only operationally feasible because IBM Safer Payments’ dual-access real-time database permits any kind of configuration and model change in full flight. These include changes that require complete re-evaluation of past transactions, new ways of profiling behavior, and even entirely new input variables.
Another benefit of the integrated purpose-built datastore is that since there is no external database to be procured, installed, administered, upgraded or patched. Neither parameters that must be set from the outside, the database is already configured for that purpose, reducing the chance of misconfigurations, and making solution implementation faster, simpler and less costly.
Want to hear more but prefer the live experience? Join my colleague, Constantin von Altrock, at IBM Think 2019 where he will discuss why IBM Safer Payments is used by leading global payment providers and how it is satisfying the needs of both modern commerce and customer expectations. Full session details below:
Real-Time Fraud Prevention for All Payment Channels
Wednesday, 5:30 PM – 6:10 PM
Session ID: 4450A Hilton Union Square, Ballroom Level | Imperial A