IBM Db2 Analytics Accelerator Version 7.1 — Performance with Greater Flexibility

Steven Astorino
IBM Data Science in Practice
5 min readAug 22, 2018

Organizations possess a wealth of valuable data on their enterprise systems. Ideally, decisions should be derived from this data as it’s generated to support today’s real-time business needs. Deriving insight from outdated data can lead to poor quality decisions and potentially harmful actions. This could allow fraud and cyber criminals to exploit that data and the systems. It could also result in lost revenue through missed sales opportunities. Similarly, the same could be said of not being able to analyze data fast enough.

IBM Db2 Analytics Accelerator for z/OS (Accelerator) can help address both of these challenges. Firstly, it enables queries to be performed on current transactional data with little or no impact to operational system performance. Secondly, the Accelerator can work to greatly improve the performance of applicable queries typically found in analytics applications.

The Accelerator is designed for high-speed analysis of enterprise data. It can help reduce operational cost and complexity by bringing the analytics closer to the data and enabling near real-time analytics on transactional data as it’s generated. The Accelerator transforms the IBM Z platform into a hybrid transaction and analytical processing (HTAP) environment. It can deliver high-speed processing for complex, resource-intensive Db2 queries to support business-critical reporting and analytic workloads.

Flexibility and Performance with Db2 Analytics Accelerator Version 7.1

The Accelerator has been available for several years and is used by IBM Z clients across many industries to help speed complex queries on business-critical data, without having to move that data to a different platform for analytics. Executing transaction and analytical workloads simultaneously on the same platform can help reduce latency while leveraging the security capabilities of the IBM Z platform. Analytical processing does not degrade transactional workload performance because transactional and analytical queries are each processed within separate compute resource pools.

As depicted in Figure #1, the Db2 Analytics Accelerator Version 7.1 delivers two deployment options, either on the IBM Integrated Analytics System (IAS) and/or directly on the IBM Z mainframe.

Figure #1: Dedicated Appliance (Accelerator on IAS) or Native on IBM Z (Accelerator on IBM Z)

Choosing the Right Option

The Db2 Analytics Accelerator for z/OS Version 7.1 provides the option to deploy the form factor that best suits an organization’s requirements:

  • Db2 Analytics Accelerator on IAS allows organizations to deploy on the IBM Integrated Analytics System appliance. This pre-configured offering includes hardware (compute, storage, and networking) and software for easy deployment, management, and high performance. It also offers security-rich, flexible and elastic data storage that is both easy to deploy and manage.
  • Db2 Analytics Accelerator on IBM Z allows organizations to deploy directly on the IBM Z mainframe, within a Secure Service Container leveraging Integrated Facility for Linux (IFL) processors. This allows for deeper integration with the IBM Z infrastructure. This option can also offer flexible capacity to respond to peak analytic workload requirements while meeting SLAs. It supports a subset of SQL, executable in Db2 for z/OS. Basing the Accelerator on Db2 Warehouse allows expanded use of the supported SQL because of improved SQL compatibility across the Db2 family.

Both deployment options are based on IBM’s analytical engine, Db2 Warehouse. This helps assure a uniform experience with one API and database engine. It allows organizations to invest once in enabling their applications and workloads for the Accelerator and then transition easily between deployment options. The IBM Db2 Warehouse provides the following advantages:

  • Improved SQL compatibility across the entire Db2 family of products.
  • Columnar technology delivering faster performance for many analytic queries than previous versions.
  • Containerized approach can facilitate faster, easier integration with other technologies and approaches.

Deployment on IAS

IBM Db2 Analytics Accelerator on IAS is a next-generation hardware appliance. It provides its components out of the box, including optimized hardware and software. This deployment option delivers a balanced, advanced, performance-optimized configuration. The hardware includes the rack, the physical servers and storage. It is based on IBM Power hardware and is optimized for price/performance. The software stack includes the Linux operating system, Docker software, Docker container and infrastructure management facilities.

Deployment on IBM Z

IBM Db2 Analytics Accelerator on IBM Z instances can be deployed quickly via a download & go approach. Deploying the Accelerator on IBM Z packages the software stack into an IBM Secure Service Container (SSC) logical partition (LPAR). This helps deliver a self-managed appliance running in a SSC LPAR that can be deployed in minutes.

Multiple Secure Service Containers can share the same IFLs. Multiple Accelerators, therefore, can share the same hardware resources. This can be beneficial for several production, and test/development instances, which could be provisioned and re-provisioned as necessary.

The Accelerator on IBM Z integrates into an organization’s IBM Z environment and leverages the known LPAR-memory and CPU management procedures. This approach also leverages the organization’s own disk to store Accelerator-side data. This option is for organizations that want to take advantage of IBM Z facilities for activities such as disaster recovery and allows for deeper integration with the IBM Z infrastructure.

Summary and Next Steps

When real-time insight matters, IBM Z, along with Db2 for z/OS and the Db2 Analytics Accelerator, can help deliver a powerful transaction and analytical environment.

It can help organizations take advantage of their enterprise data where it originates to work to drive high impact decisions while minimizing latency, simplifying infrastructure, helping to reduce the cost of such IT infrastructure and improving data security. It can help turn data into insight, and insight into opportunity to provide near real-time answers to some of the most critical questions.

To learn more about IBM Db2 Analytics Accelerator Version 7.1, click here.

Stephen Astorino,
VP of Development, Hybrid Cloud, z Analytics and Canada Lab Director
IBM Analytics

Follow me: @astorino_steven

--

--