Extreme Transaction Processing (XTP)

GigaSpaces XAP and Extreme Transaction Processing (XTP)

Linear scalability, extremely low latency, and high availability on commodity hardware


Today's mission-critical transaction processing systems-from electronic trading to online banking to e-commerce-are burdened with more data, more clients, and more services than ever before. In an environment where milliseconds matter and processing demand is growing exponentially, traditional transactional approaches are falling short when it comes to scalability, latency, and availability. These shortcomings are giving rise to the need for a new breed of Extreme Transaction Processing (XTP) applications that can:

  • Scale quickly to support fast and unpredictable growth.
  • Handle exceptionally demanding requirements in complex, distributed environments including event-driven and Service-Oriented Architectures (SOA).
  • Ensure high availability and reduce end-to-end latency to meet SLAs.

GigaSpaces eXtreme Application Platform (XAP) supports Extreme Transaction Processing with linear scalability, extremely low latency, and high availability-on commodity hardware, without changes to the application. XAP is ideal for any kind of XTP application, including:

  • Market data and trading applications in the Financial Services industry
  • Pre-paid systems in the Telecom industry
  • Online reservations sytems in the Travel industry
  • High-volume portals in Government organizations

Space-based architecture, XAP creates self-sufficient processing units that can scale out across many cost-effective computers, allowing organizations to leverage investments in current systems while benefiting from the higher utilization and flexibility afforded by SOA and other highly distributed architectures. It is easy to deploy one or more processing units on as many machines as required-with one click. Deployment doesn't become even a bit more complex when scaling from ten machines to ten thousand.


GigaSpaces XAP enables XTP applications that:

  • Deliver the lowest possible latency by accessing and processing data and events locally and in-memory
  • Collapse the tiers, thereby eliminating network hops
  • Co-locate services with shared memory
  • Scale-out without increasing complexity by adding more self-sufficient processing units (Share-Nothing Architecture)
  • Scale the data and the processing at the same time through partitioning of processing units
  • Make it easy to guarantee the reliability of each processing unit's data
  • Offer distributed transactions to coordinate complex operations across workflows

Related Resources: