Scalability Without Rewrite

Reasons to Swap Out Your Current Credit Card Solution and replace it with PaymentCardXpress® (PCX) - #3 Scalability Without Rewrite

PCX deploys a multi-threaded architecture designed to scale and meet the needs of a growing business environments. PCX runs on a Java Application Server (JAS) such as Apache Tomcat or IBM’s WebSphere. On a JAS, every transaction request runs in it’s own thread. Therefore, a correctly designed solution running on a JAS will scale to meet demand without the merchant having to change configuration options or code. For example, a correctly implemented payment card solution, running on WebSphere or Apache Tomcat will automatically scale to meet increased demand without changing the configuration or the code. For the scale correctly the following must be true:

  • Hardware - The hardware must be capable of handling the processing load. It makes sense to host a JAS on a multi-core processor. Then, not only can each core execute multiple threads, there are multiple cores available to run the threads.
  • Software - The programmer designing the payment solution must be careful to avoid unnecessary software locking mechanisms, for example mutex locks, which block threads from executing.
  • Integration – CFXWorks has observed a major inhibitor to scaling when merchants attempt to integrate their front or back-office systems using a integration technique that serializes passing transactions to and from the JAS. This can be a significant problem, especially when using pinpads. For example, an order entry individual can take many seconds, or even minutes, to respond to prompts from a pinpad. You can’t afford to serialize the interactions with one pinpad with those from other pinpads. This would prove to be a performance disaster. PCX doesn’t serialize interactions with either P2PE or EMV pinpads. PCX handles multiple pinpads concurrently.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.