Dramatically faster queries, enhanced data compression for sharply reduced storage costs, time-aware tables you can create in less than an hour—they’re all realities in the latest release of IBM DB2. These improvements and more make IBM DB2 10 and InfoSphere Warehouse 10 an even more powerful, cost-efficient, and reliable transaction-processing and data warehousing environment—as early users of the new release can attest.
New internal optimizations deliver hyper-quick query performance, which translates directly into cost savings and improved ability to meet business commitments. IBM has made numerous internal optimizations for common types of queries, joins, and aggregations. IBM engineers have also optimized access to indexes, speeding up retrieval of data. Buffer usage optimization has been improved by 75 to 80 percent, boosting overall performance and saving CPU cycles.
In addition to the optimizations, a new Row and Column Access Control feature adds enhanced security while greatly reducing the need for custom approaches to segregate data and further increasing performance. With the enhancements built into this version, DB2 delivers significant performance improvements over previous versions—right out of the box.
“We are excited to see approximately a 10x improvement in query processing performance using DB2 10 over the previous DB2 version, running on IBM System x3850 using Intel® Xeon® processor E7,” says Pauline Nist, GM Software Strategies for Intel’s Datacenter & Connected Systems Group. “Customers can now realize dramatically greater performance boost at lower cost per query running IBM DB2 10 on servers powered by Intel Xeon processors.”
Shrinking storage is another DB2 10 phenomenon. The new Adaptive Compression feature saves money for IT departments by helping to tame burgeoning data storage requirements. Data often includes repeated information such as the same city, date, or department ID. Instead of storing the same value multiple times, compression enables storing these values in a single place—a dictionary—and using a shorter symbol to refer to them.
IBM DB2 has traditionally used a table-wide compression dictionary for the entire database, which typically yields the highest compression. However, maintaining a big dictionary has an associated cost in administrators’ time. With DB2 10, IBM not only applies table-wide compression but also performs page-level compression, delivering big compression improvements over the previous release. In addition, the page-level approach eliminates the issue of time-consuming updates and maintenance thanks to the adaptive nature of the page-level dictionaries.
The improved compression means data can occupy even less space than before, which helps reduce storage requirements. The Coca-Cola Bottling Company tested the beta version of DB2 10 with the company’s SAP applications, and administrators like the results they have seen. The company’s original migration from Oracle Database to DB2 brought them a 40 percent storage savings, and a subsequent upgrade to DB2 9.7 helped saved another 17 percent. “Now, using the adaptive compression feature of DB2 10, we’ve seen an additional 20 percent, bringing our average compression savings for our databases to 77 percent, a dramatic storage savings,” says Andrew Juarez, Lead SAP Basis / DBA at Coca-Cola.
A new Time Travel Query capability acts as a data time machine for DB2 customers, offering a unique feature that sets IBM apart from the competition. DB2 users can now define the time period during which data is valid or travel through time and work with historical versions of data. “We are very excited about the new Time Travel Query feature, which will save a lot of effort in our future projects,” says Jingjie Li of the business intelligence division at LanceInfo in China.
Many organizations need to manage the time dimension of their business. For example, a reservation system must make certain that no two customers book the same hotel room or airplane seat during overlapping periods. Similarly, a bank needs to ensure that only one interest rate is valid for a loan at any given point in time.
DB2 10 enables companies to address these requirements by offering comprehensive capabilities to insert, modify, and query effective dates in the past, present, and future—which are managed as business time periods. The platform also provides optional enforcement of temporal constraints, which can help prevent problems such as overlapping hotel reservations.
Organizations have traditionally addressed the need for temporal data management by implementing homegrown, hand-coded solutions. But time-aware processing logic tends to be very complicated, hand-coded application development is costly, and performance is often below expectations. With DB2 10, the homegrown approach is no longer necessary.
Use of simple SQL statements to perform complex temporal data transformations can greatly reduce coding requirements compared to homegrown implementations. “With the Time Travel Query feature, DB2 10 greatly simplifies application code, helping reduce by at least 25-30 percent the development effort for our airline-related applications,” says Ajith Nayak of Tata Consultancy Services in India.
NoSQL technology inside DB2 allows organizations to quickly and easily deploy new applications. As companies explore the NoSQL world, they often face a dilemma. Open source NoSQL data management systems can provide the agility to make rapid-fire application changes. But relational data stores like IBM DB2 offer significant advantages that many organizations want to retain.
With IBM DB2 10, they can now have the best of both the relational world and the NoSQL world. Organizations can benefit from the easy database design and rapid development of NoSQL, together with DB2 performance, security, reliability, and recovery capabilities—as well as full atomicity, consistency, isolation, and durability (ACID) compliance.
IBM DB2 10 supports a new programming model using the NoSQL graph store paradigm with data triples. IBM has added this extra layer of capability to DB2 and taken care of all the necessary optimizations under the hood to ensure a smooth experience for DB2 users. While IBM is developing other types of NoSQL data stores for future use with DB2, the graph store technology has been released first because it is one of the most flexible and useful of the NoSQL approaches.
IBM’s own Rational software group is one of the first to embrace the new graph store capability. In fact, when the IBM Rational team evaluated the DB2 graph store against the leading open source NoSQL solutions, they found that DB2 performed up to 3.5 times better.
The DB2 10 release brings with it several other advancements. These include Multi-Temperature Data Management that allows organizations to place different portions of their data on different tiers of storage. For instance, they can put more frequently accessed data on hot storage with faster storage devices and less frequently accessed data on cold storage with slower devices. The result is higher performance, improved ability to meet service-level agreements, and lower costs as organizations extend the lifespan of their storage.
In addition, a new Continuous Ingest feature enables real-time data warehousing and operational analytics by allowing an organization to continuously feed data into a data warehouse at a high rate—even while running queries against the warehouse.
With these advancements, DB2 continues to lead the way into the future of data management. As one IBM Business Partner, LIS.TEC GmbH manager Ivo Grodtke expressed it, “IBM’s DB2 10 release is solving some of today’s great data warehouse challenges.”
Special thanks to Conor O’Mahony, who contributed to this article.
IBM big data in a minute: Bringing the power of Hadoop to the enterprise
Video: The right tool for the job
Nature of analytics video: IBM and the swan of all fears
IBM redesigns its Big Data & Analytics website with IBM Watson Foundations capabilities
Visit a website with comprehensive resources dedicated to the chief data officer role
Podcast: Learn about the InfoSphere Streams project at GitHub