In-memory OLAP

OLAP is not dead. However, new trends and requirements appear every year. To survive in such changing world every technology needs to meet those requirements, follow new trends and always be one step ahead of new changes appearance. Following these rules, OLAP changed as well. Thus, we came to In-memory BI.

Traditional OLAP 

Traditional OLAP

What do we know about OLAP in general? First of all online analytical processing is the technology about working with multidimensional data, used most often in the preparation of various reports, analysis and forecasting.

The core element of the system is a cube designed in a way to provide user with possibility to view data from different perspectives. The data in the cube has an organization using either a star or a snowflake schema. At the center of the snowflake there is a fact table that contains aggregations. Fact table is connected to a number of dimension tables containing information about the measures.

OLAP server. In the traditional three-tier data warehouse architecture, it occupies the middle tire, between the database server and the client-side OLAP tool. Besides aggregating and pre-computing the data from the relational database, it provides advanced calculation and write-back options, additional functions that extend the basic possibilities of the query languages, and other features.

Talking about OLAP types, now we have the core three:

ROLAP

Relational OLAP stores all data, including aggregations, in relational databases instead of cubes, and doesn’t use pre-computation. ROLAP works with SQL tools which work in a way of OLAP operations and send queries directly to the relational database. 

MOLAP

Multidimensional online analytical processing is the traditional and the most popular OLAP model. All data, including aggregations, is stored in multidimensional data cubes. The data in the cubes is pre-computed, which ensures fast query performance. 

HOLAP

The last but not the least, Hybrid OLAP combines features of the two previous approaches in order to provide fast query processing in combination with high scalability. In this OLAP data model, a relational database and multidimensional cubes divide the data.

How did we come to In-memory OLAP? 

How did we come to In-memory OLAP

Going deeper into the in-memory technology, let’s back to the very roots and learn what was the starting point.

After the first OLAP product release in 1992, this technology became more and more popular. And that is not a surprise. Unlike a standard relational database, OLAP allowed storing pre-aggregated results providing fast speed and user-friendly interface.

However, in the 2000s the situation changed. Relational databases caught up with OLAP technology providing the same performance as OLAP cubes. Besides, data amount increased significantly as well as OLAP computing delay.

Thus, user’s choice lean toward relational databases as planning functionality is easier to develop when connecting to RDBMS, since OLAP cubes are not designed for transactional or frequent data updates.

However, in the 2010s such systems as MOLAP are still here and there are two reasons why:

  • Provide good performance for complex KPI;
  • Rather high queries running.

But is it the limit? Is it possible to combine benefits of relational databases and OLAP? Using in-memory OLAP it becomes possible to get even more.

The 2010s brings us a breath of fresh air – in-memory OLAP technology, which is not only provides us with perfect performance but delivers fast real-time calculations. And this technology is just gaining momentum.

What is In-memory analytics? 

Let’s now making out in detail what are we talking about. In-memory OLAP is an approach where the analytical data is loaded into the memory for on-line calculations and queries. Thus, queries operation becomes faster, then in such systems as ROLAP, MOLAP and HOLAP.

All the data is in RAM, so the system does not need to access the database or physical file, which may additionally entail lots of network operations and maintain operations on the disk.

Growing and adopting 64-bit architectures in-memory analytics makes it possible to handle more memory and larger files than 32-bit, and thus, in general reducing the memory price.

Fast speed during querying becomes possible no matter KPI complexity.

Why does In-memory OLAP become so popular? 

What are those benefits that make in-memory OLAP so promising? Here are the reasons to users to choose in-memory instead of traditional OLAP:

  • Fast: unlike traditional OLAP which requires time for business reports and analysis performance, in-memory BI operation time is rather shorter. As soon as everything works in memory, the time required for analysis is significantly reduced. All relevant data is loaded into memory, thus it becomes unnecessary to connect to external data sources.
  • Multiple operation processing with high performance: 64 bit architecture allows performing more operations simultaneously no matter the complexity they have.
  • Unlimited pre-calculations: in-memory OLAP helps operate pre-calculations on the fly no matter their number.
  • Accessible: RAM is cheaper than the major traditional OLAP software that make is more affordable for every company.

Besides all these benefits, take into the account that this technology is fresh, promising and fast developing. According marketing research company LP Information over the next five years the popularity of the in-memory OLAP database will only grow.

Where to get In-memory OLAP? 

As you may know from our numerous articles dedicated to traditional OLAP, the biggest players among this technology are IBM, SAP and Microsoft. But what about in-memory analysis market?

Despite the fact that in-memory database is rather modern technology more and more vendors are ready to provide their related solutions. However, nowadays there are three solutions which are standing out:

IBM Cognos TM1Fast in-memory cube that provides budgeting, planning and forecasting in real time. This corporate planning software provides a comprehensive, dynamic environment for developing timely, reliable, and individualized forecasts and budgets.
Palo OLAP ServerAn open source OLAP-system from the German company Jedox AG. In few words Palo is:

  • open source program;
  • business analysis system;
  • decision support system;
  • multidimensional database;
  • real-time system.
QlikView by QlikTechA single, new generation business intelligence platform, based on the advanced technology of an associative data model loaded into RAM. This, allows you to analyze data without first building multidimensional OLAP cubes - a resource-intensive and expensive step, mandatory for traditional BI systems.

Besides, the In-memory analytics engine also delivers fast access to the model objects and data through reporting client applications, such as Ranet OLAP.

All main components of Ranet OLAP that work with multidimensional models fully support tabular models with In-memory database technology. No doubt, the native support of tabular and multidimensional models will allow our customers to make quick, better and less risky business decisions.

Write US