IBM DB2 with BLU Acceleration is the most sought after generation database technology which specializes for in-memory computing.
It provides an excellent performance by delivering instant insight from real-time operational data and historical data. It delivers this performance with of “load and go” setup which also does not have the constraint of other in-memory solutions.
Blu Acceleration is compatible with SQL, which leverage your existing skills for quick and easy migrations from Oracle Database.
BLU Acceleration does not require SQL or schema changes to implement and it is seamlessly integrated in DB2 10.5.It also provides business agility which enables faster time–to-value through a choice of deployment models of on-premise or thorough cloud.
Fast:
- It offers Instant insight from real-time operational data for growing revenue, reducing cost and lowering risk
- It enables to run queries more than 1400x faster
- 35x to 73x faster analytics
- The analytics is faster, 35x to 73x
- Next generation in-memory with IBM Research innovations
- This provides Next generation in-memory with IBM Research innovations
Simple:
- “load and go” performance provides simplicity in opreration.
- The reporting and transactions in the same system which enables Simplified IT landscape
- Indexes, aggregates, or tuning is not required.
Agile:
- It provides low-risk migration from Oracle Database
- It provides Extensive and simple SQL compatibility.
- It is for on-premises or via the cloud
- It provides our largest customer databases compression ranging from 7x to 20x
Advantages of BLU:
- The usability of BLU Acceleration is very simple.
- DBA's can just load and start using it.
- The absence of secondary objects, such as indexes or MQTs, is not required to improve query performance.
- Through BLU there is a good performance for analytic workloads.
Each column is stored on a separate set of pages on disk in Column-organized tables
The I/O needed for processing queries is reduced by Organizing data by column on disk. This is done as only columns that are referenced in a query need to be accessed.
analytic queries that access a large number of values from a subset of the columns and make heavy use of aggregations and joins which is favored by Columnar organization. The memory disk is loaded with only the column data which needs to be processed.
BLU provides the advantage of filtering from predicates, with significant I/O savings.
BLU does not require all the active data to be loaded in memory before it can start executing the SQL.
An efficient compression algorithm is used for Column-organized tables that are automatically compressed.
The DB2 query processing engine enables the capability to evaluate predicates and perform complex operations like joins and aggregation on compressed data. BLU helps in better query performance it also allows efficient use of memory as compressed data can be stored in memory.
The reading ranges are skipped of column values that are not needed to satisfy a query.
A small data structure called a synopsis table is utilized by Data skipping
The synopsis table does not require any configuration or maintenance to use, it is created and maintained automatically for column-organized tables
If BLU is provided by the hardware then it takes advantage of single instruction multiple data (SIMD) processing
SIMD uses the same instruction to be applied to multiple pieces of data in parallel at the hardware level.
BLU algorithms are designed to scale with the number of cores available in the system.
Massive compression, data skipping, and late materialization leads to less data, is needed in memory in order to process a query.
If a query cannot be processed entirely in memory and I/O, then BLU utilizes a memory caching algorithm that is optimized for scans to reduce the number of times specific pages are read from disk.
Regardless of table organization, your applications use the same SQL interfaces, backup and recovery scripts, and LOAD and EXPORT commands that you are familiar with as BLU Acceleration is fully integrated into DB2.
shadow table concept is introduced, from DB2 10.5 fp5. This helps to use both Columnar and Row based table in a Database.
The concept of shadow tables is critical to use OLTP and Analytics workloads.
Mostly clients insist to utilize the performance of both OLTP and Analytics workloads.