Suvarna Garge (Editor)

DB2 UDB

Updated on
Edit
Like
Comment
Share on FacebookTweet on TwitterShare on LinkedInShare on Reddit

DB2 for Linux, UNIX and Windows is a database server product developed by IBM. Sometimes called DB2 LUW for brevity, it is part of the DB2 family of database products. DB2 LUW is the "Common Server" product member of the DB2 family, designed to run on most popular operating systems. By contrast, all other DB2 products are specific to a single platform.

Contents

The latest DB2 LUW release is version 11.1.

DB2 LUW was initially called DB2 Universal Database (UDB), but over time IBM marketing started to use the same term for other database products, notably mainframe (z-Series) DB2. Thus the DB2 for Linux, UNIX and Windows moniker became necessary to distinguish the common server DB2 LUW product from single-platform DB2 products.

The current DB2 LUW product runs on multiple Linux and UNIX distributions, such as Red Hat Linux, SUSE Linux, AIX, HP/UX, and Solaris, and most Windows systems. Earlier versions also ran on OS/2.

Multiple editions are marketed for different sizes of organisation and uses. The same code base is also marketed without the DB2 name as IBM InfoSphere Warehouse edition.

History

The first release of Advanced DB2 LUW was as DB2 Universal Database version 5, available on UNIX, Windows and OS/2 platforms. This new product stemmed from two earlier products, DB2 Common Server version 2 and DB2 Parallel Edition. DB2 Universal Database version 5 continued IBM's new direction of using a common code base to support DB2 on different platforms, while incorporating the shared nothing features of DB2 Parallel Edition to support large data warehousing databases.

Key features

In addition to standard ACID-compliant row-organized relational database functionality, some of its key features are:

  • BLU-acceleration: OLAP oriented column-organized tables, compressed with order-preserving "approximate Huffman encoding", exploiting SIMD vector processing of compressed data. Because the compression is order preserving, a greater range of operations can be performed on compressed data.
  • PureScale: A data-sharing clustering of the database over multiple servers for scalability and resilience. This technology was taken from the mainframe (z-Series) DB2 product. This form of clustering suits OLTP workloads well.
  • Database-partitioning feature: A shared-nothing approach to clustering, with data hashed across multiple partitions on the same server or different processors. With the right database design, this approach allows near-linear scaling. This form of clustering is generally employed for large data warehouses rather than OLTP workloads.
  • XML support: XML-specific storage and indexing, accessible by both SQL and also XQuery.
  • NoSQL support: Currently graph triple stores and JSON support
  • References

    DB2 UDB Wikipedia


    Similar Topics