IBM i Series, Руководство пользователя

Страница: 523 / 908

Operating System/400 (5722-SS1)
493
Operating System/400 (5722-SS1)
Object Connect for iSeries (5722-SS1 Option 22)
Object Connect for iSeries provides support to simply and efficiently moves individual
objects, entire libraries, or entire Integrated File System (IFS) directories from one iSeries
server to another over a standard communications connection or over a high-speed fiber
optic bus. Systems can be connected via standard APPC (using APPN), TCP/IP
communications lines (using AnyNet), or a fiber optic bus (using OptiConnect for iSeries).
The economy of not requiring intermediate save file procedures and copies to distribution
queues saves DASD and improves performance in a manner that is non-disruptive to system
operations.
DB2 Symmetric Multiprocessing for OS/400 (5722-SS1 Option 26)
DB2 Symmetric Multiprocessing for OS/400 expands on the parallel capabilities of DB2 UDB
to improve the performance of the database for the iSeries server. This improved
performance is critical, especially in a data warehouse or decision support environment. The
performance gains allow for business decisions to be made in a timely manner.
DB2 Symmetric Multiprocessing for OS/400 further enables DB2 UDB for iSeries with
symmetric multiprocessing (SMP) on any of the iSeries N-way systems. This form of SMP
allows multiple database operations to take place simultaneously on multiple processors.
Each database operation runs on a single processor, therefore, optimizing DB2 UDB for
iSeries for online transaction processing.
With the availability of DB2 Symmetric Multiprocessing for OS/400, DB2 UDB for iSeries
becomes optimized for decision support processing. DB2 UDB for iSeries is one of the few
databases that can take full advantage of an SMP architecture for online transaction
processing environments and decision support environments.
Parallel index build capability is included in DB2 Symmetric Multiprocessing (SMP) for
iSeries. An index build can utilize multiple processors at the same time. Or in other words,
work on multiple parts of the index in parallel. This allows significant performance increases
in cases where an index is created or rebuilt. These allow faster resolution of queries
involving a join or group-by operation without the need to create and maintain indexes over
the tables involved in the query.
Both SQL and native database interfaces are supported for queries. To achieve this
parallelism with multiple processors, an individual query is split into many smaller sub-tasks.
Included in base No
Status Charged feature
Related products DB2 Multisystem for iSeries