A) Wrong doing tolerant runtime engine
The Kabira System Switchis the central web host for all Kabira applications. This can be a high performance application transaction-switch that gives the infrastructure for building and adding applications, developing networks, and mediating among data goes. It is a totally distributed program that can be deployed across multiple integrated pc processors within a single physical node, and also across multiple nodes. Additionally , the Kabira Infrastructure Move handles programmatic complexities associated with the development of programs. It frees programmers through the painstaking job of composing and reworking the code required to provide services commonly needed by simply all applications, services such as failure detection, transaction restoration, threading and concurrency, guaranteed messaging, on the net upgrade support, and target distribution. Programmers can target their focus on business requirements and creating the designs required to fulfill them. Additionally to providing a rich set of services, the Kabira Facilities Switch offers relatively small memory and disk useful resource requirements. It is because it retailers all application data in shared memory, and leverages new operating system services such as posts and other ram services, that replace efficiency traditionally furnished by middleware and database web servers. In addition it really is designed specifically for objects created by Kabira's automated code generator, so the two can perform in tandem to provide an optimum implementation of user types. Finally, Kabira application features is produced simultaneously upon UNIX (Solaris, HPUX) and Windows NT to influence the talents of each system. This reduces the problem of goods originally targeted for one system that hardly ever run as well on a secondary platform. B) Model-based advancement environment
Most Kabira applications are designed using either other graphical target modeling tools based on UML, such as Rational Rose, or textual representations of types. Using an object modeling methodology for distributed applications permits a business fabriquer to concentrate on the functional requirements of the app, not low-level implementation particulars handled by Kabira System Switch.
The advantages of this way become progressively obvious when integrating numerous different systems, each which has its own APIs, message platforms, and protocols. Kabira Infrastructure Switch signifies all external adapters because objects in the modeling instrument, and dissimilarities are invisible from the application developer. This means designers do not need to be experts in the exterior systems being integrated just like CORBA, SNMP, Java EJB, SS7, and so forth. Designers may construct a credit application using connectors already provided by Kabira and its particular partners.
C) Fully automated code era
Kabira code generation technology support total translation of object types into exe code that runs on the Kabira Infrastructure Move. 100% of the code necessary to execute the thing model is usually generated. The generator creates C++ code that is compiled and associated with libraries furnished by Kabira and partners to create an exe program. In addition to the C++ code generator offered to put into action object types, Kabira gives code generators to combine external systems with the runtime transaction-switch. These include:
CORBA customers and computers
SNMP agent generation
Java clients and EJB servers
The assembler code generators map between different target and data models maintained all of these connectors. This solves the " impedance” mismatch problem when building applications that combine different thing and info models. Fixing this problem removes the need for additional hand code. Example mappings include:
An object style and a relational info model
ASN. 1 and IDL