Borland®
Shop
Products Downloads Services Support Partners News & Events Company Community
DataSnap

MIDAS 3
White Papers
FAQ
MIDAS 3 Licensing


Order Now

Register

Support

Case Studies

News & Press
News Articles
Press Releases

Previous Versions
MIDAS 2 Product Information
MIDAS 2 Licensing Information
MIDAS Client for Java

jobs@borland.com

Got Training

We Support CORBA

 MIDAS 2 PRODUCT INFORMATION

Enterprises are Choosing MIDAS

    As computing infrastructures get larger, decentralized, and more complex, corporations are looking for the technologies that can provide solutions today. Borland's Multi-tier Distributed Applications Services (MIDAS) delivers distributed applications faster, easier, and with higher performance on bandwidth-challenged networks than any competing solution. MIDAS thin-client applications are easy to develop, easy to deploy, require zero configuration, and automatically update server based business logic without re-deploying thin-clients.

    MIDAS provides the fastest time to market for distributed applications through the seamless integration of front end thin-clients and middleware technology. Its server based business logic and database connectivity make applications responsive to changing business needs while its advanced, thin-client database architecture delivers powerful distributed applications.

    Customers use MIDAS to deliver their high-performance, scalable distributed applications.  Demand for MIDAS has been greater than anticipated in the first quarter of 1998.


Analysts See Tremendous Growth in Distributed Computing

    According to the Gartner Group, 40% of white-collar workers will be location-independent or mobile by the year 2000. Gartner Group also stated, "The percentage of worldwide IT budgets allocated to supporting bandwidth-challenged mobile and remote users will grow from less than 5 percent in 1996 to over 30 percent by 2001 ..." By 2001, IDC estimates the distributed application development market will be US$1.2 billion, while the Standish Group expects it to grow to over US$4.4 billion.

    This phenomenal growth in decentralized computing represents tremendous opportunity and need for rapid, reliable development of distributed applications. With MIDAS' distributed, scalable architecture, organizations can take full advantage of this radical shift to location-independent, mobile computing.


The explosive growth of the Internet, global computing, and the popularity of portable computers has driven a tremendous need for distributed database applications which are responsive on low-bandwidth networks. You may have a large percentage of your organization's work force physically disconnected from your corporate data for most of their working hours. You may be producing an application your customers use directly to place orders or perform queries on data you provide to them. These clients may be working on portable computers with sporadic dial-in access to your databases.

  • Clients need to be able to make changes to the data they're working with while they are not connected to the database, so they can be instantly responsive to business demands.
  • Client data queries must be fulfilled as quickly as possible, no matter what bandwidth or distributed computing transport is being used.
  • Client updates to the data need to be processed intelligently and efficiently
  • Data conflict resolution has to be robust and handled gracefully, on the client, on the server, or both.
  • Client-side enforcement of constraints on the data need to be controlled by the server, so data is validated when the user is entering it, rather than having to make a costly round-trip to the server for validation.
  • These data constraints need to be configurable on the server, with the clients automatically enforcing the new constraints whenever they connect to the server.
  • These thin-client applications need to be as thin as possible, without requiring a large footprint for data drivers.
  • You need to implement and make changes to these distributed database applications immediately, as your organization's business needs change.
  • You need to be able to add additional servers as application use grows, and automatically use those servers without re-writing the thin-client application.
  • You need automatic server fail-over to ensure an application server is always available for your client at any time.

Borland MIDAS (Multi-tier Distributed Application Services Suite) answers all these needs with unprecedented speed, scalability, and ease.

Even if you don't need distributed database applications yet, MIDAS is a better solution for your Client/Server applications. Its high performance on low-bandwidth networks, automatic implementation of intelligent cached updates and advanced conflict resolution make MIDAS superior to conventional Client/Server implementations. If you need to provide a distributed version of that Client/Server application in the future, MIDAS allows you to make it multi-tier without requiring you to rewrite any of the source code.

MIDAS provides data remoting and synchronization solutions for distributed applications that are not tied to any specific platform or distributed computing transport. MIDAS gives you access to Any Data, Any Time, Anywhere with any distributed computing standard, as the following illustration shows:

Borland MIDAS is:

  • fast, with an intelligent middleware layer which brokers data to optimize database retrieval, updates, and transmission; increase user capacity and transaction volume by generating less network packet traffic; and optimize the use of the chosen data transport by transmitting only the essential data between the application server and the thin client.
  • scalable, your database applications are instantly scalable from one user to thousands of users by simply adding more machines providing the same services. MIDAS' support for location transparency and load balancing allows you to implement fail-over, redundancy, and on-demand scalability without changing your server source code. You can even simultaneously support any of the standard distributed computing transports, like CORBA's IIOP, DCOM, or TCP/IP without requiring a change to the application server source code or a DCOM to CORBA bridge.
  • dynamic, with data packet technology which supports server-based changes to database schema, business rules, and data constraints without requiring re-deployment of thin clients access the application server. Industry-standard transports such as CORBA, DCOM, and TCP/IP can be used to transport the compact, self-describing data packets among NT-based servers and thin clients running on Windows 95, Windows 98, Windows NT, and any platform supporting Java.
  • easy, providing cross-platform multi-tier application development through a suite of advanced components, services, and core technologies. In addition to multi-tier implementations, Borland MIDAS also enhances the reliability and performance of single tier and two tier (client/server) database architectures by providing sophisticated cached update intelligence without requiring additional code

You can rely on Borland MIDAS when building responsive, scalable, maintainable distributed applications that need to be delivered to the market faster. MIDAS solutions can be developed with the Client/Server or Enterprise editions of Borland Delphi, C++ Builder, and/or JBuilder.

MIDAS Technology

Borland MIDAS provides cross-platform multi-tier application development through a suite of advanced components, services, and core technologies. MIDAS enables businesses to deploy distributed business objects with a feature set that includes:

  • Distributed datasets and transaction processing
  • Cross-platform thin-client applications and applets
  • High-speed database connectivity
  • Automatic database constraint propagation from the server to the client
  • High server availability with fail-over safety
  • Load balancing
  • Connection pooling and resource management
  • Advanced conflict resolution
  • Briefcase mode/mobile database operations

These features have been designed to work individually or in conjunction with each other, and support operating system and industry standards.

Borland MIDAS includes the technologies that provide enterprise-wide scalability for the Borland family development of tools, including Borland Delphi, C++ Builder, and JBuilder.

New in MIDAS 2

We have radically simplified distributed databases with new transports and database features, data schema optimizations, and unprecedented user interface improvements, such as:

  • A completely cross-platform and cross-product distribution solution with native CORBA support using Borland VisiBroker
  • Optimized database connectivity and server utilization through native MTS support
  • Connections to more database file formats with support for easy access to non-BDE (Borland Database Engine) datasets
  • Improved server and client communications via TCP/IP with call backs implemented in the socket server
  • Support for advanced database design and implementation with native Oracle 8 enhancements
  • Intelligent retrieval and updating of data with nested datasets. This feature implements nested data as trees of objects, so multiple levels of detail can be managed by simply connecting one dataset to another on the server. This allows details to be transferred to the client in the same datapacket the master uses. It also allows master and details to be applied in a single transaction and ensures the order of the updates is correct
  • Fail-over support for any transport with the new SimpleBroker Component
  • Data provider enhancements:
  • You can now use a provider that will work from any TDataSet descendent (TDataSetProvider), and the basic provider is no longer tied to BDE. The provider/resolver framework is now completely written in Delphi, so IDPROV32.DLL is no longer required.
  • Fine control over the update process with the BeforeUpdateRecord event (replaces TUpdateSQLProvider).
  • There are more options to specify how the SQL for the update will be generated. You can specify if a field is in the Update and/or Where clause or if it is a key field. You can specify the UpdateMode now for updates (so you can do "Where All", "Where Changed", "Where Key"). There is also a "resolve to dataset" option that allows the records to update the dataset itself. This method means that all events on the source dataset fire.
  • Transfer optional parameters to and/or from server to client
  • All relevant field properties can now be passed to the client, so that the server can specify things like DisplayName, EditMask, etc and have that propogate to the client automatically.
  • Send calculated fields from server to client
  • Control which fields get sent from server to client via the Fields Editor
  • Ability to control which fields get used in the WHERE, UPDATE and SELECT clauses of the resulting SQL
  • ClientDataset enhancements:
  • Bind parameters via a Params property
  • Easy support for briefcase mode with the new FileName property that will automatically store retrieved datasets when shutting down the client application
  • Filters have been enhanced to support most SQL functions (SUBSTRING, UPPER, etc). Filters can also search memo fields as well.
  • Data Grid enhancements support visualization of advanced new data types like ADTs, nested tables, REFs, and VARRAYs Supports nested datasets (and other Oracle8 ADT types)
  • Pop-n-drop datasets
  • Expandable and collapsible headers for ADTs
 
Site Map Search Contact