I’ve been meaning to write about SAP HANA for a while now! However, things continue being pretty busy at my end and I’m still trying to find the balance between my thesis, job interviews, meditation sessions and giving presentations at esteemed institutes – the latest being Nyenrode University! An honor to represent the World Youth Forum for Ethics in Business it was very satisfying to watch the growing interest among students on this topic and their willingness to take concrete steps here.
Anyway, as the adage goes – better late than never! I’m really short on time but it is time to pen down some thoughts about HANA.
One of the earliest “definitions” I came across for HANA (High Performance Analytic Appliance) was – It’s a column oriented, in-memory database appliance that supports processing and analysis of large amounts of data. I pick this ‘version’ of the definition since it gives us an insight into how and what HANA was positioned from the ‘early days’.
Most managers and public reports talk about the value of using HANA – results are available quickly, with better insights, hence business execution can be made faster and better.
But, I don’t want to jump into the business right away. I think it might have a stronger impact to first understand what those words in HANA’s definition, and then discuss the business implication. HANA, undoubtedly, will be a series on posts on my blog!
What does “Column Oriented” mean?
Column oriented means that all the values in a column are stored together. This is in contrast to row-oriented, in which all the values for a row are stored together. This is a very simple and straightforward example to understand “row-oriented” and “column-oriented” structures. Many operations are considered more efficient with a column-oriented approach, since they are computed over large numbers of similar data items. This columnar structure is similar to the NOSQL databases. Using this structure, HANA process billions of rows of data to reducing the time to get an answer from days to seconds. Efficient in terms of speed and space required, the column is always in-memory which means I/O is never required. This actually provides the HANA system with an advantage over disk-based systems. However, HANA supports both row and column types. This blog goes a little more in-depth about the row-vs-column debate. You can skip the read if you aren’t particularly interested in details.
What does “In-Memory” mean?
Typically In-memory is seen as the storage of information in the random access memory (RAM) rather than on disk drives. once on the RAM, it is faster to detect patterns, analyze massive data volumes ‘on-the-go’ and perform operations quickly. The drop in memory prices has had a major impact in contributing to the popularity of in-memory computing. HANA supports the running of its ERP applications, data warehouse, business intelligence systems, and related non-SAP databases, applications, and other systems on one database. HANA compresses data to a very large extent, thereby putting more data in the RAM! The column-based structure is a key component here.
What is ‘transaction processing” and “analysis of large amounts of data”?
Processing information into individual operations, such that, a response is generated immediately to the user, is called transaction processing. HANA supports both structured and unstructured data (although this is still somewhat unclear to me) and has the ability to aggregate and update large volumes of data in real-time. It takes the approach of MapReduce. For a brief description on MapReduce you could read one of my earlier posts.
So in short, to answer “What is HANA?”. It is a database and information processing platform. I found this article on the SAP website a succinct one – worth the read!
The SAP HANA Cloud Platform!
With practically the whole industry talking about Cloud and HANA being integral to SAP, it was but time before something was done about it. SAP did not disappoint! In fact they’ve done something truly fantastic! A few weeks ago SAP rolled out the HANA Cloud platform.
So what is the “SAP HANA Cloud Platform”? It is SAP’s in-memory Platform-as-a-Service offering enabling developers to build, extend, and run applications in the cloud!
Why do I think it is fantastic? Because they’ve gone deeper in the stack on the cloud. HANA is now a “platform”. It is no more just a database, but a platform capable of delivering various functionalities and capabilities. A platform for developers to build applications and increase the ‘value’ of adoption and use. This makes it more valuable – if, of course, people use it!
I have absolutely no idea about the licensing and costs. In fact in my readings and talks with people I deliberately leave this topic! SAP is confusing enough with their regular price lists – I can only imagine what must be going on here 😉 There’s much I would like to write about but perhaps another time!
For now, I would guess the next two critical steps SAP should/might consider around HANA are –
1) Keep the partner ecosystem intact. This will be key to them building enterprise applications rapidly and use the SI’s (in particular) to push HANA aggressively. Fiori has been launched. Making sure the user-interface, ‘need’ for these apps and integration issues will be a deal breaker.
2) Acquisitions – Datacenter’s are key. More private / secure cloud centers are needed. There are already questions about the entry into hardware. This is confusing – since everyone pokes fun at Oracle for the SUN acquisition and yet we have Google, Microsoft and possibly now SAP, who will advent into hardware!
I learnt a new dutch phrase today – toi-toi-toi 🙂 I know I am not doing justice, missing out on talking about so many things (broader implications of Big Data, Mobility, Fiori etc etc etc), but I am sure you will forgive me! Hope you enjoyed the read despite my hurry in writing this particular post on HANA!
P.S – Here is an update covering the gaps in this article to give a more through overview!