Analytics without Boundaries for the Future-Ready Enterprise

Running IBM DB2* on the Intel® Xeon® Scalable processors powers the next generation of business innovation.

DB2 is IBM’s flagship database product which supports in-memory column store tables for analytics workloads.

Hardware at-a-glance: Intel® Xeon® 8180 Platinum processor

Software tools/libraries: Intel® C++ Compiler

Modern businesses find themselves awash in data, and their best path to a competitive edge comes from analytics that can put that data to work in real time. An IT strategy must enable intelligent, rapid-fire decisions today and prepare for new business models and capabilities tomorrow. Building blocks from IBM and Intel are a solid foundation for ongoing innovation.

IBM DB2* running on the new Intel® Xeon® Scalable processors builds on an ongoing history of joint successes and industry leadership. Massive scalability and throughput turns the deluge of data facing every business from a challenge into an opportunity.

For well over two decades, IBM and Intel have collaborated closely to create symbiotic capabilities in their products for the enterprise. With each generation of IBM software and Intel® architecture, business customers extend their capabilities to put data to work, advancing the horizons of capability and opportunity. As business intelligence and analytics have become more central to profitability, real-time database performance has become vital.

Proven Performance for the Data-Fueled Enterprise
Testing reveals that performance increases occur when refreshing the server hardware and the DB2 database platform. For mainstream two-socket servers, refresh of hardware alone from previous-generation systems based on the Intel® Xeon® processor E5-2699 v4 to the Intel® Xeon® Platinum 8180 processor increased performance by approximately 1.51x.1

“In today’s data-driven economy, businesses require solutions that put data to the best use possible. Having the ability to make decisions real-time while planning long-term strategy is critical. That’s why the combination of IBM DB2* V11.1 and the new Intel® Xeon® Scalable processors is so compelling. This technology stack puts more power, flexibility, and insight into the hands of business decision makers.”

Dinesh Nirmal, vice president, Analytics Development, IBM Hybrid Cloud

Always-On, In-Memory Analytics with IBM DB2* Database
DB2 V11.1 is built to power the next generation of demanding, simultaneous workloads, including those for advanced analytics, cognitive, and transactional computing. Its in-memory architecture scales smoothly across large core counts in enterprise servers based on the Intel Xeon Scalable processors, contributing to improved throughput and response times for the full range of leading-edge business applications.

IBM BLU Acceleration* delivers the breakthrough performance advantages of in-memory columnar processing, taking advantage of the latest cache and memory subsystem advances of Intel architecture. By delivering rapid system response, BLU Acceleration safeguards the ability of next-generation business applications to advance real-time business insight, as well as an excellent user experience.

To help satisfy the challenges of the cognitive era, DB2 meets demanding processing requirements head-on, while embracing the full scope of data sources that today’s enterprise applications must draw on, from sensors and mobile endpoints to the data center, web, and cloud.

By loading data continuously from multiple enterprise sources, DB2 keeps data available where it is needed to support fast insight and decision making. Compression using multiple techniques on that data maximizes compression of columnar tables, helping DB2 drive performance up further. Advanced data management and manipulation techniques provide highly efficient use of hardware resources throughout the solution stack.

“The new Intel® Xeon® Scalable processors are the most compelling platform advance in a decade, and the advances built into DB2 V11.1 showcase just how valuable upgrading the whole solution stack can be. This is a golden opportunity for joint customers of IBM and Intel to get huge return on investment from combined hardware and software refresh.”

Lisa Davis, vice president and general manager, IT Modernization Group, Intel

With an eye to their futures, businesses now have a simple path to powering better decision making using vast amounts of data. By upgrading their analytics solutions to DB2 V11.1 running on server hardware based on the new Intel Xeon Scalable processors, enterprises can embrace the potential of opening their solutions to rapid operations on massive data. At the full range of system sizes ranging from two to eight sockets, the increases in throughput available from hardware and software refresh covers the gamut of solution scale.

Performance drives insight, and the competitive advantage often goes to the competitor who identifies opportunities first. In a world where those opportunities lie hidden in vast seas of data, analytics powered by IBM and Intel offer a path to success.

Informações de produto e desempenho


Workload: 300 GB BD Insights, 12 users, geomean of three runs; performance metric: queries per hour (QpH).

Baseline: 2x Intel® Xeon® processor E5-2699 v4 @ 2.2 GHz; 256 GB DDR4-2400 RAM (16x 16 GB); Intel® SSD DC S3700 Series (800 GB); Red Hat Enterprise Linux* 7.3; BIOS: 86B.0271. R00.1510301446. Raw score: 3774 QpH.

New: 2x Intel® Xeon® Platinum 8180 processor @ 2.5 GHz; 384 GB DDR4-2667 RAM (24x 16 GB); Intel SSD DC S3700 Series (800 GB); Red Hat Enterprise Linux* 7.3; BIOS: 86B.01.00.0412.020920172159. Raw score with DB2* v10.5.08: 5717 QpH. Raw score with DB2 v11.1.1.1: 6848 QpH.

2Performance results are based on testing as of May 2017 and may not reflect all publicly available security updates. See configuration disclosure for details. No product can be absolutely secure.
3DB2 v11.1.1.1. The IBM Big Data Insights Internal Heavy Multiuser Workload (BDInsights) is a multi-user data warehousing workload based on a retail environment. The workload is comprised with a mix of complex and intermediate queries. The scale factor for the workload is 300GB with 12 users. Testing by Intel® and IBM® May 2017.