Microsoft announces rebranded PDW as the Analytics Platform System

Posted by  Jen Underwood   in  ,      6 months ago     8370 Views     1 Comment  

newAPS

Earlier this week at a Microsoft customer event in San Francisco, new CEO Satya Nadella announced Microsoft’s future vision for their data and analytics platforms. Using a “data culture” theme, several executives discussed what the future roadmap looks like at a high level. In an interesting twist, PDW (Parallel Data Warehouse) was rebranded as Microsoft Analytics Platform System with additional appliance component offerings. Microsoft is pitching the Analytics Platform System as a turnkey big data analytics appliance combining MPP SQL Server data warehouse with HDInsight, Microsoft’s 100% Apache Hadoop distribution into a single solution. To integrate data from SQL Server PDW with big data from Hadoop, Analytics Platform System (aka APS) offers the PolyBase data querying technology.

For those of you that follow Microsoft, you may recall Dr. Dewitt’s truly fantastic 2011 PASS keynotes where he discussed NoSQL not replacing relational databases but rather being a complement. Well folks, here it is coming to life in 2014.

PASS2011DrDewitt
-Source Dr. Dewitt’s 2011 PASS keynote presentation

APS
-Source Microsoft APS field rep

Since I have dear friends that work with PDW, I asked for their take on this news. They are pleased by the market attention but actually are unhappy with yet another name change since PDW was beginning to be recognized and gain some momentum. Personally I like the name and think it aligns well to the analytics industry trends that I monitor out here in the wild.

Getting other, non-Microsoft biased peer perspectives, I learned that the APS solution is a nice offering but may have a little more maturing to do. Right now it uses a distributed query model that brings an entire table of data over to the APS (PDW) side before running analysis. For my Microsoft friends – if this last point is not true – please let me know and I will update it. (4/18/2014 CORRECTION: My Microsoft PDW friend, a deeply technical and highly reputable source, has confirmed that the distributed query data movement is no longer true. They now use “predicate pushdown” to Hadoop that generates map-reduce jobs behind the scenes to do the work on the Hadoop side when necessary.)

PredicatePushdown

In the world of big data analytics, solutions such as Amazon RedShift, Hadoop vNext with Stinger, Hawq, Shark and Impala are also compelling to compare with APS for similar solution architectures.

Two New World Records for TPC-H Benchmarks
At the conference, Microsoft revealed two new world records for TPC-H Data Warehousing Price/Performance benchmarks. More information on that big news is posted on the Microsoft Data Platform blog.

Microsoft’s data platform executives reiterated the recent launch of SQL Server 2014 that is a truly significant improvement in their in-memory OLTP and all up BI story – aka SAP HANA compete. The true beauty of SQL Server 2014 in-memory capabilities is the minimal to no application code changes to enjoy the added performance improvement! If you do have SQL Server apps and want high performance in-memory upgrades, seriously look no further than SQL Server 2014. The other in-memory solutions in the market do require time consuming code rewrites.

Intelligent Systems Service
An intriguing announcement was made on future Machine Learning initiatives in the cloud with a preview of a new Intelligent Systems Service – that sure sounded a bit like Splunk if I understood this offering. Executives went on to say that the Machine Learning work originated from Bing and was adapted to a deep vertical domain to democratize data into customer actionable information. This is interesting to me because it is something totally different for the Microsoft data platform…it sounds like an end-to-end vertical industry solution offering.

Still No Predictive News
I was hoping to hear something, anything really, about predictive analytics from Microsoft. I follow many vendor predictive analytic offerings and literally I am dying over here to hear about predictive from my friends over there. Come on guys, PASS BA Conference possibly?

All in all this was a big week, jam packed with quite a few announcements for my former Microsoft data platform product management family, friends and peers that I do adore. In a few more weeks, I look forward to sharing what I expect to be cool news from the May 7, 2014 PASS BA Conference keynotes.

About  

Founder and Principal Consultant, Jen Underwood has almost 20 years of hands-on experience in the data warehousing, business intelligence, reporting and predictive analytics industry. Prior to launching Impact Analytix, she was a Microsoft Global Business Intelligence Technical Product Manager responsible for technical product marketing and field readiness for a $10+ billion market suite of analytics offerings spanning across Microsoft SQL Server, Office and SharePoint. She also held roles as an Enterprise Data Platform Specialist, Tableau Technology Evangelist and a Business Intelligence Consultant for Big 4 Systems Integration firms. Throughout most of her career she has been researching, designing and implementing analytic solutions across a variety of open source, niche and enterprise vendor landscapes including Microsoft, Oracle, IBM, and SAP. As a seasoned industry presenter, author, blogger and trainer, Jen is quite active in the global technical community. Recently she was honored with a Boulder BI Brain Trust (BBBT) membership, Tableau Zen Master (MVP) award, PASS Excel BI Chapter leadership role and a Dun & Bradstreet MVP. She writes articles for BeyeNetwork, SQL Server Pro and other industry media channels. Jen holds a Bachelor of Business Administration degree from the University of Wisconsin, Milwaukee and a post graduate certificate in Computer Science - Data Mining from the University of California, San Diego.