How Accurate Was John Appleby on What Oracle Won’t Tell You About HANA?

Executive Summary

  • John Appleby made bold predictions on HANA.
  • We review how accurate he was his article on What Oracle Won’t Tell You About HANA.

Introduction

John Appleby’s article on the SAP HANA blog was titled What Oracle Won’t Tell You About HANA and was published on April 8, 2015. We review the accuracy of this article.

Our References for This Article

If you want to see our references for this article and other related Brightwork articles, see this link.

Notice of Lack of Financial Bias: We have no financial ties to SAP or any other entity mentioned in this article.

  • This is published by a research entity, not some lowbrow entity that is part of the SAP ecosystem. 
  • Second, no one paid for this article to be written, and it is not pretending to inform you while being rigged to sell you software or consulting services. Unlike nearly every other article you will find from Google on this topic, it has had no input from any company's marketing or sales department. As you are reading this article, consider how rare this is. The vast majority of information on the Internet on SAP is provided by SAP, which is filled with false claims and sleazy consulting companies and SAP consultants who will tell any lie for personal benefit. Furthermore, SAP pays off all IT analysts -- who have the same concern for accuracy as SAP. Not one of these entities will disclose their pro-SAP financial bias to their readers. 

The Quotations

SAP HANA is Less Expensive than Oracle?

“I don’t want to get into detailed pricing debates on a blog, but I help customers implement SAP HANA every day, and it’s less expensive than Oracle, period.

HANA is, in our estimation, the most expensive database in the world. We covered this in detail in the article How to Understand S/4HANA and HANA Pricing. HANA is the only database priced per GB. Oracle is probably the second most expensive database after HANA, but there is no way that HANA is less costly than Oracle unless the database is kept extremely small. HANA has higher long term overhead than Oracle because even in 2019, HANA is still plagued by maturity problems, which we covered in the article Why Are Some Customers Removing HANA?

This quotation demonstrates that as is usually the case, Appleby is unwilling to provide evidence for his contentions.

HANA is Available at a Lower % of Your SAP Software Estate?

“Firstly, it’s available at a lower % of your SAP software estate than Oracle – so you will actually get a license payback if you implement HANA. Sure, that’s different to ROI, but it’s a nice place to start.”

We do not know what this means. Appleby could be far more precise when promoting a benefit.

Business Suite on HANA?

For any apps where you don’t want to implement HANA, SAP throw in SAP ASE included in the price. And remember that if you buy S/4HANA in 2015, you will get all the future innovation included for the price of the SAP Business Suite on HANA runtime license last year.

This is also quite confusingly written. But first, runtime licenses are not production licenses. Therefore they should never be used to estimate costs. They are deliberately used by SAP to lull customers into underestimating costs. They were used by Forrester to incorrectly reduce the costs of HANA, as we covered in the article What Oracle Won’t Tell You About HANA

Appleby is describing the inclusion of ASE, or the Adaptive Server Enterprise, which is picked up in the Sybase acquisition. This makes the estimation of the true cost of HANA far more complicated, which is why SAP does this.

Like for Like Application Platform License Cost?

Second, if you compare like-like for an application platform license cost, HANA comes out less expensive. The data reduction of HANA’s compression and simplified data models, easy to buy appliances and flexible license packages mean you can license HANA for less money than an equivalent Oracle database including all the features you need. Because SAP don’t charge extra for HA/DR, you are more likely to build a resilient solution!

Neither Oracle nor HANA is an application platform. They are databases as we covered in What Oracle Won’t Tell You About HANA, but if we switch to merely calling them databases, Appleby is again incorrect.

Based on HANA have many column-oriented tables it compresses, but not anywhere near as much as SAP has stated, as we covered in the article The Problems with the Strange Lenovo HANA Benchmark.

As we covered in Does S/4HANA Have a Simplified Data Model?, HANA does not simplify the data model of the application.

It is unclear that HANA even has an HA/DR capability, so SAP may not charge for it, but it is also in no way comparable to Oracle’s HA/DR. Therefore this is a poor comparison.

SAP Bundles Many Items with HANA?

Third, if you want any kind of Enterprise solution, SAP HANA is available at a single cost for all features, including the equivalents of Enterprise, Multitenant, RAC, Active Data Guard, Partitioning, Advanced Compression, Advanced Security, OLAP, Advanced Analytics, Spatial and Graph, Database In-Memory, Diagnostics Pack, Tuning Pack, Database Lifecycle Management Pack and more. If you want to get the best out of Oracle, the cost is simply crippling.

The cost of RAC for a full-rack Exadata X5-2 is over $3m alone (don’t forget the extra $1m for the spare Exadata!). For this reason, most customers don’t use advanced Oracle functionality – it’s just too expensive.”

Appleby is moving directly into HANA sales with this quotation. Oracle does “nickel and dime” customers with every new addition to an already expensive database. Just for the record, we think Oracle’s database is far too expensive. But what Appleby is leaving out is the only the base price for HANA is exorbitant and is amazingly, even more costly than Oracle. Secondly, many of the items listed here by Appleby aren’t used in HANA customers as they just aren’t competitive.

SAP on Oracle Has No Future

“This may be controversial, but in my opinion, SAP on Oracle has no future. Take a look at the April 2015 Roadmap Update if you want to see what I mean.

For example, if you put your SAP data workload on Oracle 12c, you aren’t currently able to update it (yes, seriously!). You can’t use any of the innovations Oracle invested in including Oracle Database 12c In-Memory, Oracle Database 12c Multitenant, Oracle Database 12c ADO/ILM or Hybrid Columnar Compression for Exadata. And that’s just the operational limitations.

In addition, SAP on Oracle will only be supported until 2025, and this has serious implications for SAP customers, many of which operate a 5-10 year roadmap. Systems are expensive to implement and maintain, and so customers need to plan ahead.

The replacement to SAP Business Suite 7.0, SAP S/4HANA, will only run on the HANA platform. SAP on Oracle has no future.

If Appleby had to do it over again, he probably would not make this statement. Oracle is the predominant database used on SAP accounts. HANA has replaced very few Oracle instances as a percentage of the Oracle instances are used under SAP applications. This is 2019, and SAP stopped reporting customer numbers on HANA in the second quarter of 2015 as we covered in the article Why Did SAP Stop Reporting HANA Numbers After 2015?

Oracle had already added column-oriented tables and “in memory” when Appleby wrote this article.

Notice how far back Oracle and IBM add column-oriented tables and “in memory.” 

Appleby is jumping around a lot in his other assertions. First, only ECC has a support limit of 2025, and therefore ERP on Oracle. This does not extend to other SAP applications, and we predict this support deadline will be moved because it has to be as such a small percentage of customers will be on S/4HANA by 2025. Appleby’s quote here is no only FUD, but it was also wildly inaccurate.

Appleby receives a Golden Pinocchio for his claim that SAP is finished on Oracle. 

SAP HANA Enables a New Class of Applications?

“One of the things which is great about working with HANA is that innovation comes seamlessly and often. There was a time when HANA updates came too often, but that pace has slowed over the last 18 months with two major releases a year, verified for datacenter usage. There are some more frequent updates for early adopters or those testing new functionality, but for most, it’s now possible to get great innovation every 6 months with minimal disruption.

HANA SPS09 came in November (soon to be datacenter verified) and it provided a ton of innovation including integrated ETL and cleansing, Multitenancy, support for Big Data and IoT, Event Stream Processing and much more. Read What’s New in SAP HANA SPS09 for more details. SPS10 will be on us very soon, and promises to continue that trend.

Oracle 12c has been out for 2 years and is only now even supported for SAP applications, and very few innovations have come even since Oracle 11g.

What this means is that HANA enables a new class of business application to be developed. In S/4HANA Logistics, SAP achieve over a 70% data reduction by removing redundancy. What happens if we now mix in spatial awareness of product and customers, how might we enable predictive maintenance and ensure stock is proactively in the right place?

I can tell you one thing for sure: you ain’t seen nothing yet!”

Most of what SAP calls “innovation” in HANA has been due to SAP having to reverse engineer other databases as we cover in Did SAP Just Invent the Wheel with HANA?

That is, HANA has seen rapid turnover in code but is not breaking any new ground in databases, obviously as it is copying items from other databases. Even in 2019, HANA has all kinds of missing functionality. This is functionality that already exists in other mature databases. Oracle has far fewer changes, but Oracle is far more mature than HANA, so does not need to add them. As few customers leverage anywhere close to what Oracle has already, stability is far more important to Oracle.

SAP does not reduce redundancy by 70%, as we covered previously. The rest of the quote is just meaningless.

No Hardware Lock-In?

“At the time of writing, there are 538 configurations in the Certified SAP HANA Hardware Directory and this number increases every day. You can run HANA on anything from a shared VMWare appliance with just a 64GB slither of memory, right the way up to a 112TB cluster, and beyond. There are 22 Certified Enterprise Storage solutions if you want to reuse an existing hardware asset.

What’s more you can keep your vendor honest because hardware configurations are very easily compared, so you can decide what provides you with the right value for money and functionality.

By contrast, Oracle locks you in to 1/4, 1/2 and full rack solutions which you can see on the Oracle Engineered Systems Price List. Their $1m full-rack solution has a paltry 2TB of DRAM – that’s what you get in a $100k pizza box from any one of 11 SAP HANA vendors. And that SAP HANA 2TB appliance is big enough (and mission-critical enough) to run almost any SAP ERP system out there, apart from the very large enterprise.

Want HA? No problems, that will be another appliance and 2 clicks of the mouse to configure. No costly RAC licenses and complicated implementation.”

The sales feel to Appleby’s writing is again quite apparent. Appleby is presenting himself as independent of SAP, but this does not at all read like a person with any independence on this topic. And of course, Appleby is primarily employed to charge billing hours for himself and other Bluefin Solutions consultants (at the time this article was published).

Oracle would love to sell customers high-end Exadata servers to run Oracle on, as is listed in the Oracle Engineered System Price List, and we agree that for most applications, these are not ethical values. However, Appleby makes it sound as if Oracle requires customers to run Oracle on Exadata — which is not true. Most Oracle databases in the world are not run on Exadata hardware.

HANA HA can be configured with two clicks on a HANA appliance? That is quite unlikely. Again, HANA HA is not even known to be effective.

SAP HANA is a Cloud Platform?

“SAP HANA was always envisioned as a cloud platform, despite its beginnings as an analytics database. The HANA Cloud Platform brings the capabilities of a full-scale cloud application platform for developers and ISVs alike.

The HANA Enterprise Cloud brings managed-cloud-as-a-services capabilities for businesses who would previously buy on-premises databases, and includes a partnership with IBM SoftLayer.

SAP S/4HANA will be available in the public or private cloud, depending on your requirements – or indeed on-premise. One thing is worth noting: S/4HANA is cloud-first, and on-premise second. Cloud users are first-class citizens in S/4-land.

This doesn’t diminish from the fact that many enterprise software customers still want to put their database on-premise – that will continue, especially in regulated industries for the foreseeable future, but it is a nod to the fact that cloud software is the future, even in the Enterprise.”

HANA is a database and never had anything to do with the HANA Cloud Platform, or now known as the SAP Cloud. This is genuinely wholly false. SAP Cloud has very few people developing for it, and it is not an application platform. One would have to be deranged to highly financially biased to try to develop using SAP Cloud.

The IBM SoftLayer partnership did not work because the Softlayer acquisition by IBM failed, as we covered in the article AWS and Google Cloud’s Competitive Capabilities and the IBM Softlayer Acquisition.

S/4HANA is almost entirely run from on premises. SAP has virtually no public cloud customers of any kind. S/4HANA is on-premises not only first, but almost entirely.

Appleby receives a second Golden Pinocchio Award for the relationship between HANA and HANA Cloud Platform.

Final Words

“When Steve wrote his blog in 2012, the FUD from Oracle was focused around how HANA wasn’t mission critical, how it didn’t work and didn’t scale. Looking back at this in 2015 it’s clear that HANA has come through that FUD unscathed, and Steve’s points were spot on – in fact just as valid as they were back then, which is why I didn’t repeat them.

What’s even more interesting is how Oracle has stood still since 2012 whilst HANA has delivered more and more capability. Now, hundreds of the world’s biggest businesses are running their transactional systems on SAP HANA and new business applications like S/4HANA have arrived that do things that Oracle cannot do.

The conversation we are having around the Oracle FUD is distinctly different from the one we had in 2012. Now, Oracle are saying that SAP HANA isn’t actually as fast as Oracle (benchmark it yourself if you doubt that), that SAP doesn’t understand the cloud and that SAP on Oracle is the smart choice.

I look forward eagerly to revisiting this conversation in 2018 and seeing how the world has changed once again. My sense is we will be living in a very different world then!”

As a shill for SAP writing an article like this and many others so filled with falsehoods, Appleby should be legally restricted from using the term “FUD.” Steve’s article was not only entirely false. It is a cringeworthy piece written by a person with what appears to be no database knowledge, as we covered in the article Analysis of Steve Lucas’ Article on What Oracle Won’t Tell You About HANA.

The middle portion or second paragraph of this quote is entirely false. HANA has continued maturity problems with HANA appearing to go in circles and unable to meet its marketing literature. In 2019 and according to DB Engines, HANA is roughly 4% as widely used as Oracle. SAP’s firing of a large component of the HANA development team in March of 2019 as we covered in SAP’s Layoffs and a Brightwork Warning on HANA illustrates that even SAP has tired of its massive investments in HANA.

We have many data points from various customers telling us that HANA underperforms for Oracle as we covered in, What is the Actual HANA Performance?

The performance of HANA is not only slower than Oracle, but it is also slower than IBM DB2 and most likely SQL Server.

Appleby talks about revisiting the world in 2018, well it is 2019, and nothing that Appleby has stated has come true.

Comment 1: From Stefan Koehler

“You can’t use any of the innovations Oracle invested in including Oracle Database 12c In-Memory, Oracle Database 12c Multitenant, Oracle Database 12c ADO/ILM or Hybrid Columnar Compression for Exadata. And that’s just the operational limitations.”

Stefan’s quote below is responding to Appleby’s quote above.

“Sure you can. Oracle In-Memory will be supported at the end of Q2/2015 (SAPnote #2133079) – that is only 3 month after the Oracle 12c certification for SAP. Some features were excluded in the first certification cycle (Oracle 12c was officially certified at End of March 2015 first), but this does not mean that you can not make any use of them at all. The certification of all the mentioned features will be rolled out in cycles as promised. However i guess you forgot the “operational limits” of HANA like database restore on a smaller (or bigger) system. No simple database copy anymore from PRD to DEV or QA – same hardware infrastructure (nodes) needed for all the system landscape levels ;-))”

This is in a similar spirit to what we pointed out earlier but focused on different areas.

“By contrast, Oracle locks you in to 1/4, 1/2 and full rack solutions which you can see on the Oracle Engineered Systems Price List.”

Stefan’s quote below is in response to this quote above.

“I don’t get this argument as you compare Oracle Exadata platform with SAP HANA, but you should compare SAP HANA with Oracle 12c + In-Memory option. However you have no vendor or OS lock at all with Oracle 12c + In-Memory option. In contrast to that SAP HANA has some kind of OS locking (Linux: SLES/RHEL).”

This is correct.

First, Appleby misstates that Oracle is locked into Exadata (it is if you purchase Exadata, but most Oracle customers do not). Then Appleby makes his point by comparing HANA against the wrong item. And there is no lock-in to the hardware with Oracle 12c + In Memory.

“Want HA? No problems, that will be another appliance and 2 clicks of the mouse to configure. No costly RAC licenses and complicated implementation.”

Stefan is responding to this Appleby quote above.

“Sure because SAP HANA has nothing like RAC at all (see previously mentioned point + reference).”

This is also true.

Both Stefan and Appleby have more comments, but we found them less interesting than this comment from Stefan.

“Please don’t get me wrong. SAP HANA is really good at performance topics (e.g. BI workload – much better in parts than Oracle, etc.), but i am pretty tired of all the other “false” comparison. I also know your passion about HANA (from SCN) and you are really good at SAP HANA of course – no doubt, but i can not understand how this blog post gets an average rating of 4 and above with such kind of misleading and partly wrong information. This is also the problem i see “out-in-the-field” – the SAP HANA information is consumed unfiltered/unproved by customers. I think the SAP marketing guys should get some Oracle training first and then they would see that some statements / comparisons are just not true. Or even better – why not stop comparing and start trying to close-up in some core functionality and let the customers decide based on their own freedom of database choice. (To me) It seems like SAP is frightened, because of their SAP HANA sales output (6,000 customers on HANA within 5 years is not that much i guess as measured by the marketing effort) – so they choose to go the “panic support way”. By the way i know some of these 6.000 customers and most of them are just “play-ground” systems related to size, load and users … and yes i know there are also some big players on SAP HANA … i just wanted to reveal the number 6.000 :-))”

Stefan is quite right. Appleby has filled this article with false comparisons.

Stefan is also correct that 6,000 customers on HANA versus the marketing output put into HANA is not that high. And HANA grew little from when Stefan wrote this comment, and it is now four years after.

Conclusion

This article receives a 1 out of 10 for accuracy. It is filled with false comparisons and predictions that did not come true. Appleby functions in an undifferentiated manner from an SAP salesperson. It is difficult to differentiate what is written in this article from what an SAP sales rep would say to customers.