|
|
This section contains information on the following subjects:
WebLogic Real Time provides lightweight, front-office infrastructure for low latency, event-driven applications. For companies in highly competitive environments where performance is key and every millisecond counts, WebLogic Real Time provides the first Java-based real-time computing infrastructure.
For example, for certain types of applications, particularly in the Telecom and Finance industries, stringent requirements are placed on transaction latency. When these applications are written in Java, the unpredictable pause times caused by garbage collection can have a profound and potentially harmful affect on this latency.
For this reason, WebLogic Real Time's proprietary JRockit JVM introduces "deterministic" garbage collection, a dynamic garbage collection priority that ensures extremely short pause times and limits the total number of those pauses within a prescribed window. Such short pauses can greatly lessen the lessen the impact of the deterministic garbage collection when compared to running a normal garbage collection.
WebLogic Real Time includes the following components:
| Note: | For WebLogic Real Time 1.0, this requires acquiring a special JRockit license, as described in " Installing and Updating License Files" in the WebLogic Real Time Installation Guide. |
For more information about the key components of the WebLogic Real Time, please refer to Architecture.
For a listing of the hardware and software configurations supported by WebLogic Real Time, see WebLogic Real Time 1.0 Supported Configurations.
These use cases provide examples of how WebLogic Real Time can provide solutions for high-performance environments with response-time sensitive applications.
An investment arm of a large retail bank provides an exchange for derivatives of European securities. It is an over-the-counter (OTC) request-for-quote and execution system (but provides no settlement and clearing services). A broker submits a request for a quotation and includes the investment identifier and quantity. The system accepts the quotation and applies certain business rules. Depending upon the investment identifier and market conditions, the request is routed to a particular third-party market-maker who then calculates and provides the bid and ask price for the derivative. The response is returned to the broker via the OTC exchange. The broker can then execute the trade of the derivative through a subsequent request, which is routed via the OTC exchange to the appropriate market maker.
The complication with this arrangement is that arbitrage traders can take advantage of the latency delay in the bank's OTC exchange infrastructure because the arbitrage trader can measure the latency that occurs during the small period in which the request for quotation is handled. In a fast moving market, price changes of the derivative may occur within this latency period. This presents an opportunity for an arbitrage trader to take advantage of inefficiency in the marketplace and expose the investment bank to intolerable risk.
The investment bank requires a very high performance-driven software infrastructure, such as WebLogic Real Time. It requires that the latency of the OTC exchange be extremely low. Specifically, to combat arbitrage traders, the latency of the exchange's infrastructure must be less than the latency of the arbitrage traders' infrastructure. In this way, the arbitrage traders' data becomes stale before the exchange's, and therefore is not actionable.
A large investment bank is a market-maker for fixed income securities. A request-for-quote (RFQ) is received from an inter-dealer market electronic communication network (ECN), such as TradeWeb. This RFQ would have been submitted to a number of entities. To be competitive, it is vital that the quotation is returned as quickly as possible with the best possible price. Therefore, a minimum amount of latency is necessary to ensure that the investment bank wins customers, or at least, the latency is less than that of the organization's competitors.
During the quotation process, a risk and pricing model is executed to determine the quote price to provide to the customer. Because of the complexity of these calculations, they are currently performed overnight. The result is a stratum of at least four grades of risk advisories that govern fixed rate securities prices. Note that there is at least a twelve-hour lag in these risk calculations. This leads to a risk window since the calculations are stale even at the start of next-day business. To lower this risk, and potentially provide better rates to customers, a real-time risk and pricing calculator would be required. WebLogic Real Time provides a latency-adverse infrastructure to make this feasible.
WebLogic Real Time 1.0 is certified to be compatible with the following platforms and software:
Table 1-1 lists the operating system and hardware configurations on which BEA supports WebLogic Real Time 1.0.
Table 1-2 lists the software supported by WebLogic Real Time 1.0.
You are entitled to support if you have a support agreement with BEA.
Table 1-3 defines the terms and acronyms used this document:
|