Big News @BigBlue for Big Data in the Cloud

[ With thanks to my co-author and fellow technology executive in the cloud, Adam Kocoloski ]

Great people make awesome things happen.  Over the past week IBM has announced several major extensions to its cloud technology platform for data management. IBM's leading data warehouse in the cloud, dashDB, has added shared nothing cluster support (Massively Parallel Processors - MPP), and now scales from megabytes to petabytes, along with major extensions for Oracle compatibility, Twitter integration, R, and a host of ISV tools. In parallel IBM announced the acquisition of Compose, now adding MongoDB, PostgreSQL, Redis, RethinkDB, and Elasticsearch to the cloud portfolio. A super spread of powerful technology with major cloud computing benefits: No IT to manage, easy, fast, and low cost.  Enough of the marketing... let's talk tech.

IBM's dashDB Poised for Power

dashDB is a cloud-based data warehouse providing in-memory performance with load-and-go simplicity. No configuration, no tuning. It's based on cutting edge in-memory vector processing for maximum performance with Business Intelligence applications like Tableau and Cognos, and for processing R scripts and machine learning for data scientists. New this month is the extension of IBM's powerful in-memory data warehouse technology to an MPP scalable architecture with super scalability.  This new architecture offers the benefits of faster than RAM in-memory speed, from the vector processing architecture, and an architecture that has nearly unlimited scale-out to an arbitrary number of servers and cores.  This means that dashDB can be used for small megabyte scale data sets up to massive volumes measured in hundreds or thousands of terabytes. In fact in recent tests we found data compression in dashDB to be factors better than other leading cloud providers.

The processing power of this scalable MPP architecture combined with in-memory optimized algorithms is breathtaking.

“dashDB reduced data queries that used to take all day down to four seconds, totally transforming our business. IBM customized a solution for us, and provided greater support than competing offerings.” - Thered10

Polyglot Data Warehouse: dashDB Speaks Many Programming Languages

As a premier data warehouse from IBM, dashDB naturally is DB2 language compliant, so all of the SQL, PL SQL procedures, JDBC, and ODBC applications written for DB2 will run on dashDB.  However, this new technology is now also Oracle enabled, adding a wide range of Oracle syntax support for JDBC, PL/SQL procedures, Oracle specific SQL nuances, OCI, and even SQL*Plus scripts.  The team decided to go all-in on this, so the technology is implemented as native language support in the dashDB kernel and not as an emulation layer. That means you can now run Oracle applications faster than Oracle, by running then on dashDB and getting the benefits of the in-memory columnar vector processing technology.

Bring Your Fav BI Tools

dashDB is growing its support for BI tools every day, but chances are your favorite BI tools already work well. Cognos, Tableau, Microstrategy, Looker, Aginity, SAS, R-Studio, Esri ArcMap and of course IBM's natural language cognitive computing technology Watson Analytics. Fuelled by dashDB's in-memory columnar processing these BI tools can now extend their reach into new domains of performance and applicability.   

Five Paradigms to Leverage dashDB

1. Standalone data warehouse or data mart in the cloud. dashDB is a very effective replacement for data warehouses and data marts you would have created on-site in the past. Now faster, easier, and lower cost, you can leverage the cloud to get more done.
2. Development and QA Systems. Many companies are not able to place their data in the cloud for legal and policy reasons. Even so, dashDB offers a superb low cost and easy way to do development and test for data applications, simplifying the infrastructure your team builds on so they can focus on their work and not on the IT and database infrastructure.
3. Augmenting the existing data warehouse (Hybrid). Not all data is equally important and you can easily use to cloud to store colder data that is infrequently accessed. This reduces you IT capital costs and operational costs as well.
4. Analysis of NoSQL Data. dashDB is seamlessly integrated with Cloudant - IBM's NoSQL JSON store. With a push of a button from the Cloudant dashboard you can create a dashDB warehouse populated and continuously synchronized with Cloudant. Part of the beauty of this technology is the automatic schema discovery that goes on automatically to map your JSON data directly into a schema in the data warehouse.
5. Data science data store. dashDB isn't just a data warehouse for BI. It provides rich in-database analytics for R, spatial analytics and predictive modeling. These analytic functions run in the memory space of the warehouse which dramatically speeds their execution. (Copying all the data out to another processing engine is what normally makes analytic engines slow).

Here's dashDB in use with R-Studio

Here's some cool spatial analytics with dashDB:

IBM Acquires Compose adding MongoDB, PostgreSQL, Redis, RethinkDB, and Elasticsearch to its Cloud Portfolio

IBM Aquired Compose, a great company that built their business operating MongoDB, PostgreSQL, Redis, RethinkDB, and Elasticsearch over time. Their services will be the perfect complement to IBM Cloudant’s Apache CouchDB-based service and to IBM’s cloud data warehouse, dashDB, among other Bluemix services.

Compose designed its service to run in the public cloud, but containerization caused the Compose team to gain a unique understanding of the full stack involved in its service. To the point where Compose now also manages deployments running in data center colocation facilities around the world.

It speaks to the maturity of the company Compose has built. From a data management perspective, they know how to use the cloud to get capacity online rapidly, but also how to optimize infrastructure to deliver a premium experience for customers and their applications.

• Information Week IBM Acquires Database-as-a-Service Company Compose https://ubm.io/1LNRFb0
• Vator News IBM acquires database as a service company Compose https://bit.ly/1fAcmuD
• Tech News Today International Business Machines Corp. Invests To Acquire Cloud-based Startup Compose https://bit.ly/1SJJa10
• Bidness ETC International Business Machines Corp. To Acquire Database Startup Compose https://bit.ly/1LOhUQa
• Celebrity Cafe IBM steps up open-source cloud commitment to developers https://bit.ly/1IC43M0