Print Save PDF

About 4 minutes

The explosive growth of both structured and unstructured data from multiple sources requires businesses to derive insights faster than ever from massive amounts of data just to keep pace. Users want and need modern applications that store, retrieve, and discover value from these new data sources. This is the essence of big data analytics.

Developers need to create new applications that leverage new, rapidly changing unstructured data types – often across more than one database. In addition, many organizations want to evaluate Open Source Database software to reduce the costs and restrictions associated with traditional commercial Relational Database Software (e.g., Oracle DB license fees).

Power Systems Designed from the Ground Up for Big Data

To support this broad range of next-generation, data-centric solutions, IBM Power Systems now offers POWER8 servers and solutions optimized for today’s most challenging big data environments. IBM POWER8 is designed for the most challenging big data and analytics environments.

POWER8 has system features that result in 4x the threads per core, 4x the memory bandwidth, and much larger processor cache per socket when compared to x86 Systems technology. In addition, POWER8 incorporates innovation from a growing, open ecosystem of next-generation database technologies that can broaden application choices, lower TCO, and deliver valuable insights much faster.

Wide Range of Applications to Meet Your Needs

In the taxonomy of the modern DB world, we tend to divide databases into Relational or NoSQL choices.  There’s usually no one-size-fits-all database for everyone’s needs. More commonly, clients and developers will need to take advantage of the data stored and available in multiple databases. Data from everything ranging from a single line of business application to enterprise-wide applications would be included.

Relational Database Management technology products are optimized for transactional integrity and take advantage of the large ecosystem around the SQL language and standards. IBM POWER Systems Linux environments offer a very broad and diverse range of Relational Database Management solutions.

Relational Databases

Currently available commercial and open source Relational Databases on Linux on Power include: DB2 and Informix from IBM, SAP HANA, SAP IQ and the SAP Adaptive Server from SAP, PostgreSQL from Enterprise DB, MariaDB, and open source versions of MySQL, SQLite, Hive, PrestoDB, Derby, Firebird, and the Percona DB Server.

NoSQL Databases

NoSQL databases are much lower cost and can provide clients a simpler data model with more dynamic control over storage and retrieval of primarily semi-structured or unstructured data types.  

Below is a brief inventory of 4 flavors of NoSQL databases that are all available today and in some cases deeply optimized on Linux on IBM POWER8:

  • Key/Value Stores run on POWER8 Linux and include Redis, Helium, Memchaced, Risak KV, Level DB, ZODB, RocksDB, WiredTiger, and Tokyo Cabinet. Both Redis and Helium exploit IBM POWER8 CAPI FLASH technology for greatly enhanced performance.
  • Document Stores that are available on Linux on Power include MongoDB.  
  • Columnar Store NoSQL DBs available on Linux on Power include Cassandra, Hbase, and Accumulo. Cassandra also exploits IBM CAPI Flash on Power Systems for environments that needed greatly enhanced performance results. 
  • Graph Store database technology on Linux on Power includes Neo4J.

Open Source Search

Open source Search technology is also available on Linux on POWER via products including both ElasticSearch and Sphinx. IBM BigInsights is IBM’s optimized Hadoop and Spark based environment for a more robust, performant and lower TCO big data analytics set of solutions for Linux on Power.

Try running your next big data solution on Linux on IBM POWER8. The combination of IBM POWER8 with Linux and either Relational or NoSQL Databases enables and delivers fast results for real-time decision-making. They can also better handle large volumes of data, improve systems performance, and lower TCO.

Hardware is often an overlooked component, but it is critical to your database performance. Get your NoSQL Developers’ Guide to Infrastructure and Maximizing Data Performance.

[Click Below] 


Written by Chuck Calio

IBM Power Systems Growth Solutions Specialist