In-Memory First
SAP achieved a true transaction and analytical (i.e. translytical) platform by designing SAP HANA from the ground up with a new data processing architecture in mind. This architecture greatly simplifies applications built on this platform removing many of the legacy overhead required to optimize applications on traditional disk-based database solutions.
SAP HANA is the only true in-memory columnar data store optimized for both high-speed transactional scenarios and also very fast query performance optimally supporting both OLTP and OLAP application scenarios. With SAP HANA, there is a single copy of data as there is no need to move data to external systems to offload analytic workloads or populate materialized views and in-memory caches to achieve acceptable performance. This allows applications built on SAP HANA to process transactional and analytical workloads at high speed without any data duplication enabling businesses to have accurate insights based on the latest data.
Other solutions in the market typically provide an in-memory “option” to existing disk-based, row-oriented solutions, requiring a copy of data to be made into in-memory storage. Row-based storages don’t yield the same level of data compression as columnar storages leading to higher costs and inefficiencies. These solutions may provide some performance improvements in certain use-cases, but they do not deliver the overall value that SAP HANA does by truly reducing the data footprint, increasing overall performance for all data and reducing management costs of the solution.
Managing Data of All Sizes
Does it always make sense to have all data in-memory? Realistically, this is not necessary. SAP HANA provides an innovative way to manage hot and warm data storage in an optimal way to strike the right balance between price and performance.
SAP HANA Dynamic Data Tiering provides multistore table partitions which easily allow a single table to span across memory and disk. This greatly simplifies access to hot and warm data in consuming applications and greatly reduces the administrative effort to manage large data volumes. This provides for much lower total cost of ownership by reducing the amount of hardware needed, lower administration costs, and more efficient IT operations overall.
Foundation for a Modern Data Platform and the Intelligent Enterprise
SAP HANA is a modern multi-tenant in-memory platform to build next-generation Big Data, IoT, translytical, and advanced analytics applications. SAP HANA allows advanced analytics using a single data platform. How is this done?
SAP HANA natively provides capabilities for machine learning, spatial processing, graph, streaming analytics, time series, text analytics/search, and cognitive services all within the same platform. Applications on SAP HANA can provide real-time actionable insights from a wide-variety of data types.
Combining these innovative technologies with SAP HANA’s advanced application development, modeling and custom algorithm support inside the database layer, more business logic can be executed close to the data. Specifically, SAP HANA’s support technology like stored procedures, business functions, core data services, rules framework, calculation views, and the application function library provide developers with a way to easily model complex data-intensive operations as close to the data as possible. Most competitive solutions require that you copy the data between core data storage and other technology stacks to perform these advanced analytics on the data.
View the full post on www.approyo.com
No comments:
Post a Comment