Create an ODATA service with HANA and R

Interesting example of wrapping R up into an exposed ODATA layer out of SAP HANA.
http://scn.sap.com/community/developer-center/hana/blog/2013/10/08/creating-an-odata-service-using-r

How about using SQL in R?  Lubridate'ing?  Random Forests?
http://blog.yhathq.com/posts/10-R-packages-I-wish-I-knew-about-earlier.html

Exporting to Excel? Plus installing a bunch of other packages in a single shot?
https://gist.github.com/bearloga/10988512

Boasting about Oracle's In-Memory Database

Oracle is trying to halt some of the migrations from Oracle to SAP HANA with their new Oracle 12c In Memory technology.  Basically it allows you to "Pin" tables in memory in a columnstore cache.

Some speed boasts:

• Database queries and analytics running between 100 and 1,000 times faster than in the past.

• With in-memory technology, Oracle 12c database allows each CPU core to scan 2.5 billion rows per second.

• The time it takes for the 12c database to process 10 million invoice lines has been shrunk from 244 minutes to 4 seconds.

• The time it takes to run a financial analysis program is cut from about four hours to roughly 12 seconds

• A system for keeping track of a company’s transportation network featuring 16,000 drivers and 60 million shipment data records, is slashed to under a second from 16 minutes.

• A process that had previously taken 58 hours now needs only 13 minutes.

Welcome to a world where disks are a thing of the past.  Pretty soon I predict that physical disks will go the way of tape drives, and we'll all be running with 2-4 terabytes of RAM.

http://diginomica.com/2014/06/11/larry-ellison-database-fast-customers-broken/?utm_source=dlvr.it&utm_medium=twitter&utm_campaign=Feed%3A+diginomica2+%28diginomica%29

SAP HANA SPS 08 is out

Lots of new features and fixes for SPS 08 can be found in SCN or on Twitter.

SAP & the new SQL 2014 Cardinality Estimator

If you're running SAP on Microsoft and are lucky enough to be current on your SQL environment, you already know SQL 2014 was released yesterday and have probably been testing the CTP for months.  Right?

Anyhow, performance and compatibility are the two areas most likely to cause issues, or pleasant results.

Some more info on the new SQL 2014 Query Optimizer.
http://blogs.msdn.com/b/saponsqlserver/archive/2014/01/16/new-functionality-in-sql-server-2014-part-2-new-cardinality-estimation.aspx

If you're running, or upgrading to Oracle 12c, changes to their DB Optimizer with Adaptive Plans could affect your performance. 

http://scn.sap.com/community/oracle/blog/2014/02/19/oracle-db-optimizer-part-x--looking-under-the-hood-of-adaptive-query-optimization-adaptive-statistics--sql-plan-directives-oracle-12c

While testing Dynamic Sampling at my previous project, I noticed that whether dropping / recreating indexes, updating statistics, or using Dynamic Sampling, you always paid for your performance someplace.  As data gets larger, the built-in scheduled maintenance jobs can no longer cope with updating large partitioned tables in a 4-hour time window.  Custom solutions may need to be implemented.

Both SQL 2014 and Oracle 12c have enhancements to partitioning strategies.  Oracle can now delay the global index maintenance on an entire table when a partition is modified.  Truncation and exchange changes can be cascaded through referenced partitioned tables.  Interval partitioning is available. Partial indexing to speed up bulk loads.

For SQL 2014, online maintenance of index partitions and lock priorities seems to be one of the biggest feature improvements for partitioning.  Incremental creation of statistics on partitioned tables could also help with availability.

Getting started with Microsoft BI, Hadoop & SAP BI

Some links to get started with some of the various vendor technologies that integrate with Hadoop.

Azure HDInsight with Hadoop 2
http://www.windowsazure.com/en-us/solutions/big-data/

PowerQuery for Excel 2010/2013-- connect to Hadoop
http://www.microsoft.com/en-us/download/details.aspx?id=39379


Try out SAP HANA
http://www.saphana.com/community/try


Try out SAP Lumira or use free at home
http://www.saphana.com/community/learn/solutions/sap-lumira


Hadoop on a VM
http://edataanalyst.com/2013/08/hands-on-hadoop-tutorial-with-hortonworks-sandbox-vm-part-1-the-boss-edition/


Some articles around Oracle & Hadoop & tying it all together
http://blog.gbrueckl.at/2013/11/sap-hanas-big-data-scenario-with-power-bi/
http://www.oracle.com/technetwork/database/bi-datawarehousing/twp-integrating-hadoop-data-with-or-130063.pdf


Oracle vs. Couchbase - from Oracle
http://www.oracle.com/technetwork/products/nosqldb/documentation/nosql-vs-couchbase-1961718.pdf

Oracle vs. Couchbase -from the Comparison Wiki
http://vschart.com/compare/oracle-coherence/vs/couchbase

Next Hadoop (THUG) meetup in Toronto with Don Pinto from Couchbase.

http://www.meetup.com/TorontoHUG/events/169706762/

HANA Deep Dive

So you want to learn more about HANA?  A couple places to start are from last year's deep dive sessions.  Unfortunately I missed the pre-conference sessions at TechEd 2013.  However, a lot of good tidbits were garnered from many other HANA sessions.

The TechEd sessions are here.

Watch the deep dives from HANA Distinguished Engineers.

Follow the Engineers on SCN.

Many of the SAP roadmaps are out and all roads lead to HANA.
SAP IQ Roadmap here.

Comparison between SAP IQ & HANA.