9 Questions To Ask When Choosing An In-memory Database

The digital economy is all about speed and agility. Organizations of every size need the ability to sense, respond, learn, adapt, and predict in order to meet and create customer demand in the moment of opportunity. The trouble is, traditional 20th century databases are simply unable to cope with the demands of the 21st century business. The typical model of multiple, specialized databases has resulted in a highly complex IT landscape that hinders business agility and focuses resources on maintaining existing systems rather than innovation.

A true in-memory database that combines transactions and analytics on a single platform can help you meet all of these challenges. IDC estimates that an in-memory database can deliver performance improvements of 10x for transactions and up to 100x for analytics. It can also dramatically simplify your data model and streamline your IT landscape, reducing total RDBMS costs (including server, storage, network administration, database administration, and database-related operations) by 40-45%.


But you need to be careful when choosing your in-memory solution. Some vendors merely add in-memory cache to a disk-based database instead of designing it to be 100% in-memory from the ground up. Here are nine questions to ask when choosing your in-memory database.

  1. Are my applications accelerated without manual intervention and tuning?

With disk-based databases only pre-selected data is copied in the appropriate in-memory cache, leading to extensive testing and tuning. With a true in-memory database, applications are automatically accelerated because the data they need is readily available in real time without DBA intervention.  

  1. Can I achieve predictable response times for ad hoc queries?

With disk-based databases, only queries that exclusively access data can return results in a predictable time. With a true in-memory database, all queries – planned and ad hoc – have real-time access to the data they need and can return results within a predicted timeframe.

  1. Can I get the full picture of my business in real time?

Disk-based databases have to duplicate data on in-memory caches and create data structures for data on disk before users can perform drill-downs. A true in-memory database enables users to analyze data at any level of granularity in a self-service manner and obtain results in real time.

  1. Can I answer all my data discovery questions without adding more DBAs to do data tuning?

With disk-based databases, more questions mean more time and potentially poorer performance. With a true in-memory database, no data copies, indexes, pre-aggregates or materialized views are required to deliver real-time performance.

  1. Can my application provide analytics while updating the same copy of data in real time?

As disk access is much slower than memory access, disk-based databases will inevitably experience delays. A true in-memory database can execute transactional and analytical workloads in parallel while preserving data integrity and system performance.

  1. Am I able to run my business in real time, with all my data in-memory, ready for processing?

With a disk-based database, business innovation is limited by the speed of the disk and latency can be mitigated but not eliminated. In a true in-memory database, all data is in a compressed, columnar format to maximize access speed.

  1. Can I simplify my IT landscape with an in-memory solution?

Adding in-memory caches to a disk-based database will merely add to IT infrastructure complexity. A true in-memory solution simplifies the IT landscape by delivering application, database and integration services in a single platform. 

  1. Can I run transactions and analytics on the same system without adding more DRAM and CPU resources?

Combining a disk-based database with in-memory caches adds an extra technology layer and increases the system administration burden. A true in-memory database can take advantage of the latest hardware innovations and requires only one copy of data.

  1.  Does my database vendor have an in-memory database designed specifically for both transactions and analytics?

Combining a disk-based database with in-memory caches adds an extra technology.
Designed specifically for small and midsized enterprises, SAP HANA Edge edition, advanced version, provides the right answer to all these questions. It is based on the market-leading SAP HANA platform that is built from the ground up on a simplified architecture using in-memory technology. This allows transactional, operational and analytics workloads to be combined on a single platform, streamlining your data landscape and reducing your total cost of ownership.
SAP HANA Edge edition, advanced version, is available exclusively through SAP partners.
To find out more, https://www.applexus.com/services/sap/S4HANA  

About The Author

Renjith Thampi's picture
Renjith Thampi
Lead Consultant
Renjith is an SAP Certified ABAP consultant with hands-on experience in projects across Retail, Oil & Gas, Consumer Products, Fashion, and Apparel and Footwear industries.

Leave a Reply