- Fortune covered SAP Vora.
- We analyze the accuracy of this article.
On Sept 1, 2015 Fortune published an article titled A Look at HANA, SAP pitches Vora to bridge the big data gap.
In this article, we will evaluate the accuracy of this Fortune article.
Why the World Needs Vora
The world has been drowning in talk about big data, the massive troves of information churned out by sensors, engines, and other machinery. That information can be very useful to businesses but there’s been a divide between that often-formless data and the more structured, traditional data that resides in a company’s databases, inventory, or sales systems.
SAP (SAP, +0.70%) proposes to bridge that divide with Vora, an in-memory query processor that plugs into Spark, open-source software that developers and data scientists use to ask questions of all that data.
Apache Spark is open source (free) technology geared to speed up data queries of unstructured data, but the goal of Vora is to augment, not displace, Spark said Steve Lucas, president of SAP’s Platform Products Group. Vora, slated to ship this month, proposes to speed up queries to a company’s various “data lakes” he told Fortune.
What Fortune is not bringing up is that it is quite unclear what the value is of Vora over spark.
Secondly, HANA has a very small footprint in Big Data. AWS, for instance, does not offer HANA as part of its Big Data offering. AWS offers Spark for in-memory caching and optimized execution. One can create Spark clusters from the AWS Management Consol, but not Vora and not HANA. And it is not as if AWS does not offer HANA. But they don’t offer it as part of their Big Data offering. AWS does offer Vora, but not part of their main AWS offering. Why?
How Common is HANA and Vora Discussed in Big Data Circles
Outside of SAP marketing and sales cycles, Vora and HANA are not discussed when it comes to Big Data.
Hadoop is an open source database that is a great value and has many tools that work extremely well without the proprietary and highly expensive HANA database.
A big part of the product’s appeal will be that it plugs into both Hadoop/Spark ecosystems and into transactional data sources, including SAP HANA. “We embracing Hadoop and Spark and bringing the online transaction processing world together with them,” Lucas said.
Lucas is known to provide inaccurate information on SAP, so his credibility is low due to this history. Lucas also knows very little about databases. This is made clear in the article Analysis of Steve Lucas’ Article on What Oracle Won’t Tell Your About HANA. And once again, he makes the preposterous statement that SAP embraces Hadoop and Spark. SAP would have to wouldn’t they as Hadoop and Spark are the industry standard in Big Data and SAP is virtually nowhere with Big Data. And when Steve Lucas states that OLTP is brought together with them, it makes absolutely no sense.
OLTP has nothing to do with Big Data!
HANA is OLAP, not OLTP, so it is normally just a good practice to ignore Steve Lucas. Some people don’t really make any effort to learn the topic areas in which they work. And amazingly, Fortune simply allowed this statement to be published without questioning its obvious inaccuracy.
Vora Works With What?
Why the name? Vora was selected because it’s the Latin root for “voracious,” the implication being that Vora can consume large amounts of data, according to an SAP spokeswoman.
To be clear, the use of SAP HANA, the focal point of the company’s software push, is something SAP would recommend, but is not required. “We think Vora works well without HANA, but even better (natch!) with HANA, ” he said.
This is a strange statement. What else would Vora work with?
SAP’s Vora plugs into existing Apache Ambari console so developers can keep using their tools of choice.
SAP, a leader in enterprise software, is addressing a key need of big companies that want to query both their existing data warehouses and Hadoop data, ” said Nick Heudecker, research director at Gartner.
That is strange because that is not what Spark is used for.
The Problem with Vora and the Dominance of Open Source Big Data Products
“SAP was smart to build it on Spark which is the loudest parade in town right now and very programmer focused,” Neudecker added.
One potential downside to Vora is that lot of the programmers in this field have an affinity for open-source software and SAP, is definitively a commercial software company which means it likes to be paid for its software. It will make a free developer version of Vora available on Amazon (AMZN, +0.11%) Web Services, but it cannot be deployed in production. Otherwise, commercial-use Vora will be priced on a subscription model with an 18-month term.
Yes, that is a massive downside.
And an even bigger downside is that it is entirely unclear how Vora adds any value over Spark. And the Big Data market is dominated by open source databases and tools, which looks bad for SAP’s entry has SAP’s software cost and TCO is normally the highest in any application category in which SAP has an offering.
Making Hadoop a “Corporate Database”
IDC research vice president Carl Olofson said Vora will let companies optimize their Hortonworks (HDP, +0.83%) Hadoop and make it more like a corporate database in terms of queries and query performance.
That statement is illogical. Does Carl mean that it will make Hadoop more like an RDBMS? If so, that is not a desirable end state. I have never heard of the term corporate database before, and it is not a distinction I am aware of.
Other tech vendors are working on federated data query across different data platforms, but Olofson said the most direct competitors to Vora would be from data analytics companies like Platfora and Zaloni.
Fortune receives a score of 1 out of 10 on the article. Fortune simply allows representatives from SAP to say whatever they like. The article presents the fact that SAP is offering Vora, but does not analyze Vora or validate anything that SAP says. This article appears to be a paid placement and written by SAP.
Search Our Other HANA Content
Financial Bias Disclosure
This article and no other article on the Brightwork website is paid for by a software vendor, including Oracle and SAP. Brightwork does offer competitive intelligence work to vendors as part of its business, but no published research or articles are written with any financial consideration. As part of Brightwork’s commitment to publishing independent, unbiased research, the company’s business model is driven by consulting services; no paid media placements are accepted.