Last week, SAP announced that its in-memory HANA database platform will tightly tie together transactions, analytics and text processing within a single platform. HANA is SAP’s relatively new platform that will drive essentially everything SAP does going forward. It is a super-fast and enormously robust platform that SAP has attracted a lot of attention with, and has been able to gain a large number of initial customers for since it was released about a year ago.

The company is now in the process of moving HANA to the next level, and is releasing support pack stack 5 (SPS5), which will power transactional and analytical applications from within one platform. SAP HANA will also collapse the architectural layers between application and data processing. The company will converge database services for transaction, analytics, text and predictive processing together with application services for the application server, planning, rules and event stream processing within one in-memory based architecture.

The implication here is that these services will now be able to increase the overall speed at which results can be produced, and creates greater overall efficiency within what we believe to be an already highly efficient platform. The goal for the new features being introduced is to works towards eliminating the inherent latency and complexity that exist in today's application architecture. Converging data and application processing all within one in-memory platform architecture will help companies operate in near-real time environments, in turn gaining insights on substantial events – including dealing with big data in real time - as these vents happen.

Key features in SAP HANA SPS5 will include:

• Single environment for transactional (OLTP) and analytical (OLAP) data processing: SAP HANA will be both a read-optimized column store ideally suited for OLAP applications and a write-optimized row store suited for OLTP applications. Bringing together OLTP/OLAP is central to delivering on the efficiency noted earlier.
• Natural language-based text analytics: By expanding advanced text search with natural language processing, organizations will be able to process text and social sentiment data within SAP HANA to filter files, identify entities and analyze sentiment. HANA will provide support across 31 languages.
• Expanded predictive analytics for SAP HANA: SAP HANA will continue to expand on embedded predictive algorithms to enable enterprises to perform segmentation and clustering analyses. New predictive analysis library functions for SAP HANA SPS5 will include classification, clustering, outlier detection and data preparation algorithms. In addition, SAP HANA will also support predictive model markup language (PMML) for the integration and sharing of predictive models.
• Expanded extended application services for SAP HANA: New extended application services in SAP HANA will deliver native application server capabilities as a service. This will greatly aid developers in quickly developing and deploying applications directly within SAP HANA using HTLM5, Javascript, SQLScript, XML/A, JSON and ODATA. These services can, in turn, be exposed for consumption by device-based applications or by other application servers.
• Enhanced business rules management: Business rules management becomes a core component of the extended application services built into SAP HANA. This will empower developers to easily add business rules to their applications without embedding them directly within software codes.
• Application functional library framework: Developers will be able to easily access embedded algorithms and business functions through extended application service capabilities via SQLScript or JavaScript. This new level of support makes it easier for developers to build applications that embed advanced analytics.

Real-Time Data Platform Features

In addition to delivering SPS5 for HANA, SAP will also add real-time stream data processing and enhanced enterprise information management (EIM) capabilities to the platform. Doing so will drive additional data processing consolidation, with the goal again being to deliver accurate, real-time business insights.

These new capabilities are based, in some cases, on taking advantage of SAP Sybase technologies. New features here will include the following:

• Real-time stream data processing support: High velocity, streaming data typically produced by machines, financial exchanges and massive websites will be able to be processed in real time using the SAP Sybase Event Stream Processor (SAP Sybase ESP). Businesses will be able to store events identified by SAP Sybase ESP directly into SAP HANA at high speed, for both storage and additional analysis. Additionally, SAP Sybase ESP will be able to take advantage of the real-time analytic features of SAP HANA to evaluate complex decisions based on deep multi-sourced data in SAP HANA.
• Common modeling environment: This will operate across data stores via SAP Sybase PowerDesigner, and will enable enterprises to develop logical information views regardless of where data is stored.
• High-performance bulk loading: Bulk loading of data into SAP HANA or SAP Sybase IQ from a variety of new data sources including Hadoop via SAP Data Services software is planned to be available.
• Enhanced data profiling user interface: Will enable business users to manage metadata as business terms, monitor data quality scorecards and identify data quality issues across the enterprise — no matter where the data resides — via SAP Information Steward software.

Large-Scale Data Center Deployment


Image courtesy of Shutterstock

With SAP HANA SPS5, SAP plans to expand support for SAP HANA deployments in enterprise data centers for 24/7 operation. This effort will include the following:

• High availability of SAP HANA with hot and warm standby servers.
• Integrations to third-party backup tools will be expanded to deliver synchronized backup for improved disaster tolerance, with the option to fail over to a remote data center.
• Security enhancements such as encryption to protect data at rest and enhanced authentication, access authorization and audit logging capabilities.
• To help lower the cost of SAP HANA deployment, multiple instances of the SAP HANA database are intended to be deployed in single SAP HANA appliance for development, testing and sandboxing use.

Open Ecosystem Co-Innovation

SAP continues to buildon and expand its collection of partner programs. For SAP HANA the partner ecosystem now has more than 150 startups in the SAP Startup Focus program, and eight independent software vendors (ISVs) are building innovation on the SAP HANA platform. SAP will introduce various third-party certification programs to foster BI, ETL and back-up tools to integrate with SAP HANA in order to provide wider user enablement, data sourcing and reuse of existing investments.

Certification will help ensure that the third-party tools adhere to the SAP HANA connectivity, security, supported structure requirements and take advantage of innovative functionalities for SAP HANA. In addition, client interfaces such as ODBC and JDBC will be open for customers and partners to create applications and solutions on SAP HANA.

SAP has published a very interesting white paper that details the performance of SAP HANA conducting queries against one petabyte (one followed by a whopping 15 zeroes if you don’t know) of enterprise data containing 10 years of transactional history holding 1.2 trillion records. Overall the performance described is quite impressive even if it is SAP that sponsored the paper. The white paper is titled "SAP HANA One Petabyte  Performance."