Business transaction performance

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Business transaction performance (BTP) is a methodology providing for comprehensive visibility into complex, multi-tier business transactions. BTP provides real-time discovery of transaction topology, behavior and performance; visualization, analysis, automation, reporting and process improvement.


The goal of BTP is to identify and remediate business transaction latency in order to enable enterprises to squeeze the stealth waste out of their business and perform at the highest levels possible.[1]

BTP correlates operational, transactional and business performance real-time data onto a single pane of glass enabling users or automation to resolve latency or operational issues in a priority most favorable to the business. Users benefit with lower costs as this approach obviates the standard practice of over provisioning hardware to accommodate performance, improvements in service due to its multi-tier visibility and better management of risk as its predictive capabilities prevent business process disruption.

BTP identifies latency, and most importantly provides the operational root cause enabling its automatic or manual remediation. For BTP to be effective, deep introspection of the middleware message layer [2] is essential. Most process orchestration tools run at this level and the business processes they invoke spawn the transactions that pass between production applications. Unless the data being harvested for analysis is derived from a deep view into middleware, its vision is limited.[3]

The BTP approach enables the specification of what “business normal” and “business abnormal” states are. BTP methodology determines in real-time the behavior of transactions it is observing, including: activities that were not intended by its developers and lack of sufficient resources due to operational constraints. This means it is instantly able to compare what it has discovered, correlated and analyzed to the user description of business normal and from this it is able to “predict” business problems before users are affected; thus, enabling its practitioners to prevent problems rather than repair them.[4]

BTP can be best enabled by including a complex event processing engine [5] that is able to process millions of rules per second in order to handle the loads of the largest, most dynamic environments and find the patterns that correspond to business normal or business abnormal conditions.

The push for BTP software is due to several important changes in IT, competitive pressures and the economy:

  • Speed is money, especially in financial service where latency is measured in nanoseconds (see Low latency (capital markets))
  • Improved visibility is essential in a post-economic downturn era where increased governance, regulation and government ownership is becoming common. (see Information Technology Governance)
  • The mainframe.[4] has had a renaissance and is an essential part of business transaction environments and thus support for complex, multi-tier transactions that include the mainframe are requisite (see Mainframe Computer)


  1. ^ Gina Roos (2009). "Nastel pursues Business Transaction Performance savings".
  2. ^ Message-oriented middleware
  3. ^ Trudy Walsh GCN (2009). "Nastel’s TransactionWorks tracks hidden costs of business". Government Computer News
  4. ^ a b IBM (2009). "Nastel uses IBM Innovation Center resources to extend its value for System z users". IBM Innovation Center
  5. ^ Forrester Research (12/2010) Evaluating Innovative I&O Solutions: Converged Application Performance Management

See also[edit]