SAP HANA continues to be an unescapable phenomena at SAP SAPPHIRE where speakers, booths, advertising and coffee-machine conversations are all reminding us of the central prominence this now has within SAP and in turn for our customers. I have spent much of my time at the conference talking to representatives from SAP about the technology and the plans for its evolution. I am excited to share the highlights with you.
Firstly, the next major release will be SP5 and is scheduled for general availability towards the end of this year. In parallel to the rampant development of the on-premise edition, we now also have SAP HANA in the Cloud edition. This is available through Amazon Web services and is a real breakthrough in providing an on demand, elastic solution which is both instantly and economically available to customers for fully-fledged production usage. The latter is a key point and not be confused with its predecessor which was limited to development workloads. It has some limitations amongst which is the size limit of 62GB but a compelling proposition nonetheless. Expect further announcements around virtualisaton of SAP HANA in the coming weeks.
Returning to SP5, we can expect a number of enhancements as part of this and the following sections will provide an appreciation of this along with my thoughts.
OData will become a supported protocol, making it easier to consume data in clients such as Microsoft Office. This is a significant development and extends the reach of SAP HANA into the enterprise and in particular allows PowerPivot developers to seamlessly consume data through this interface in much the way that they can currently with other data sources. Given the prolific use of Microsoft Office within businesses, this will help deliver pervasive information experiences and help increase the SAP footprint in what may traditionally have been Microsoft oriented IT landscapes. Similarly, we are also seeing a trend towards BI as a service and this is prevalent in industries where customers want to share information with their customers, usually in B2B scenarios and the recipient would like this information in the form of a data feed which they can then incorporate within their Data Warehouse implementations. The publication of this data as a web service would make this an effortless task.
There has been further simplification through merging the existing Business Function Language BFL and Predictive Analytics Library PAL into one single, Application Function Language AFL. Yet another TLA to remember unfortunately, but hopefully allows us to forget the other two. I hope to have access to the SP5 build very soon and will examine if this has introduced any additional capability.
There was also a mention of Text Analysis in this next release and this was being discussed synonymously with the current functionality available in Data Services 4.0. I am making a slight leap here but SP5 appears to make this available within the database engine and therefore significantly ramping up the support for unstructured data; be it the identification of entities in respect to whether they are names of individuals, places, products through to analysing the sentiments being expressed. This would be much welcomed and open up enormous possibilities in wading through valuable stores of unstructured data and have this happening directly inside the database engine and without incurring data latency as would currently be the case through using the Text Analysis Transform in Data Services. Unstructured data in both on-premise and off-premise formats represents huge opportunity for businesses and is often neglected due to technical constraints that have plighted businesses to date. This opens up incredible possibilities to respond to social media expressions and manage your brand and loyalty in a way that has never been possible. Increasingly, unstructured data is being monitored and analysed alongside structured and quantitative data to provide a consistent interpretation and appropriately contextualised information.
Multiple instances on a single appliance will be supported in the case of non-productive environments. This simplifies the creation of Development and Test environments. SAP HANA One is also a candidate for consideration here. The provisioning of this has been simplified through the use of configuration wizards. This provides much needed clarity and allows customers to effectively and more economically maintain discipline around Development, QA and Production practices. I expect further developments in this area through support of virtualisation.
The Backup functionality is currently handled from the Studio Client. This will be extended through integration with third party tools. This is yet another stride to ensure that SAP HANA is truly enterprise-ready as it removes any hurdles with complying with backup and recovery procedures and it can now seamlessly fit in with the wider IT strategy. On a similar note, the integration with SAP Solution Manager has been enhanced by removing the need to connect the SAP HANA system to the internet to download and apply updates. You can now maintain the SAP HANA system through SAP Solution Manager. On the subject of maintenance and recoverability, SP5 introduces the concept of Warm standby which enhances the current disk-level replication capability. The new replication functionality will take place at the SAP HANA Data Store level resulting in data being committed to memory and hence reducing the recovery time in the event of failure.
Event Stream Processing ESP which allows the monitoring and capturing of Complex Events has been more closely integrated with SAP HANA, with this now being a supported destination. ESP is typically used in situations where we encounter a data source which is highly transactional and frequent in nature and examples of this include sensors, machine outputs, etc. The supplication of SAP HANA will result in dramatic reduction in latency, be that latency in the form of data, analysis or action. Users will have instantaneous access to granular data collated through the ESP network and the ability to analyse this through familiar and consistent formats via the SAP BusinessObjects suite.
Other improvements include usability improvements to SLT and a rearrangement of the SAP HANA Studio. The client tool has been redesigned to simplify the creation of Attribute and Analytic Views through merging this process and also introduce features such as code debugging and intelli-sense to help developers build scripts. This provides a mature development environment which users have come to expect and removes the need for the multiple steps currently required for debugging and setting traces…phew!
Lastly and possibly most significantly, the next release boasts of a new capability knows as Extended Services XS. This provides web serving capabilities built within SAP HANA which again simplifies application architectures. The current Information Composer will be amongst the first SAP applications to leverage this in what will be a HTML5 application supporting text search and an improved set of visualisations. SAP HANA continues to evolve into a comprehensive Platform which goes well beyond “just being a faster database”. Increasingly SAP HANA is delegating processing from what would traditionally have taken place in the application tier to the database tier and this is another example of this. I intend to provide additional detail on this in the coming weeks.
At risk of squeezing the next one in, Geospatial support will also be introduced in the next release. I haven’t been able to ascertain the extent of this but expect this to provide geocode support and query functions such as Points, Lines, Polygons and Buffers. Again, more on this as it becomes known.
Signing off to start learning about Mobile Business Objects in Sybase Unwired Platform…