 Major Technology 1. DSMS(Data Stream Management System)
ALTIBASE Data Stream processes the data stream based on the processing method of DSMS.
The data in the ubiquitous age share the common characteristics that they are continuously created at very rapid rate,
and the time and the sequence of creation are important. Such data is called data stream.
The appearance and the generalization of data stream have necessitated the change in the data management method in a corporation.
‘Continuous’ and ‘sequential’ data stream should be processed with DSMS, not DBMS.
It is because DBMS is the management system for the statically stored and the stationary table records.
< Comparison of Processing Methods in DBMS and DSMS >
Major Technology 2. CQL(Continuous Query Language)
Developed to process data stream, CQL performs the same role with SQL in DBMS,
but it is fundamentally different from SQL in the aspect that the queries are pre-registered in the system.
Data stream is processed simultaneously with its generation, so no further effort is required for storing the data,
or loading the data every time.
CQL has three major characteristics: one-way communication, volatility, and real-time statistics using Window.
CQL works in a one-way basis, it is impossible to load the stream again after it has been processed.
That is, the data processed with CQL is deleted right after displaying the processed result,
so that sufficient system resources can be left for use. Of course, even the processed data can be stored in DBMS,
just as in the existing systems, if it is required to store them on a short- or a long-term basis.
Major Technology 3. CEP(Complex Event Processing)
The goal of DSMS is to process individual stream in real time,
and the goal of CEP is to process in combination the relation or the pattern of multiple events generated in data stream.
CEP is the heart of EDA (Event Driven Architecture). It presents smarter and more to-the-point conclusion required for making decisions.
Major Technology 4. Coherency Table
Coherency table maintains the current status for the PK-based data record, generated in the network,
so the prompt information on the current status of an object in a distributed environment can be accessible.
This works in the same way with the replication between the tables.
Coherency table has the following advantages :
It can access the latest information on the object, to which stream processing cannot get access.
It takes snapshots of the array of corresponding objects.
It is similar in form to the record set of general DBMS, and the ODBC development can be the same with the development of database.
The identical coherency table is maintained as the same, regardless of its physical location.
DDS is the industry standard network-based technology, adopted by OMG in 2003,
for the real-time data transfer in the distributed environment.
In DDS, the format of data packet transfer is defined by the standard protocol, RTPS, and Altibase has developed its own ARTP based on RTPS.
RTPS enables Automatic Discovery of Nodes and Multicast Support, two of the major features of ALTIBASE Data Stream.
Especially, DDS enables safe and fast end-to-end data transfer.
Its de-centralized structure ensures higher reliability in the event of errors, and safe and fast end-to-end data transfer.
All the features of DDS can be utilized in the form of API,
and it transfers the data in binary format which does not have any specific data type.
Major Technology 6. Product Architecture
ALTIBASE Data Stream, composed of three feature groups - real-time data delivery, real-time data analysis and processing,
and data sharing for coherency - consists of another 6 layers.
< ALTIBASE Data Stream Architecture >
◈ Common Domain
OS Layer: This layer shows the heterogeneous operating system supplied by
ALTIBASE Data Stream. Currently, ALTIBASE Data Stream supports most of the operating systems, including Linux, Windows, HP_UX, AIX, DEC,
and SOLARIS.
CORE Layer: This layer is located on top of O/S Layer,
and is responsible for the abstraction of the operating system environment. The existence of this CORE Layer allows
ALTIBASE Data Stream to be operated regardless of the operating system, and to be ported conveniently to anywhere.
NS Layer: This layer provides the protocol of DDS,
the foundation of the features in ALTIBASE Data Stream. ARTP (ALTIBASE Real-time Transfer Protocol)
provides powerful real-time data transfer mechanism based on DDS.
◈ Processing Domain
ALTIBASE Data Stream Queue Layer:
This layer provides the data received from NS Layer, so the tasks such as input, modification,
deletion and analysis can be performed using query language. With this feature,
users can process real-time data efficiently using SQL language the same with DBMS, not using the function-style interface.
ALTIBASE Data Stream CEP (Complex Event Processing) Layer :
Coherency layer plays the role of enabling the nodes in a distributed environment to share specific data.
Using this feature, users can store the data, that should be shared in real time, in the form of tables,
and using the functions such as input, modification and query, they can share related data in real time.
ALTIBASE Data Stream Coherency Layer:
Coherency layer plays the role of enabling the nodes in a distributed environment to share specific data.
Using this feature, users can store the data, that should be shared in real time,
in the form of tables, and using the functions such as input, modification and query, they can share related data in real time.
ALTIBASE Data Stream Queue Layer: Provides real-time data
communication interface to the users in the distributed environment using ARTP features of NS Layer.
◈ Transfer Domain
ALTIBASE Data Stream Queue Layer:
Provides real-time data communication interface to the users in the distributed environment using ARTP features of NS Layer.
|
