How to Understand Exadata and Exadata Cloud Services

Executive Summary

  • Oracle has made a major commitment to appliances after their acquisition of Sun.
  • Find out how much sense these appliances like Exadata make to use.


There are four different hardware types/modalities. These are:

  • On-premises proprietary servers (which tend to be purchased from major brands like HP, Dell, Lenovo and Cisco)
  • Cloud commodity servers (which are normally run by cloud service entities and customized by those entities)
  • Mainframes
  • Appliances

The appliance is a combination of hardware and software to run a specific type of processing. One of the most prominent examples of this is Exadata by Oracle. Exadata provides an important case study of how proprietary software and hardware are combined in one hardware modality.

What is Exadata?

Exadata is part of Oracle’s engineered systems family. In particular, the Exadata (database machine) is a sealed appliance. It’s a big box similar to a mainframe. A typical Exadata X2-8 configuration comes with a price tag of $9 million. Over $6 million of that is Oracle software. According to iDatalabs, they were able to find Exadata being used by 3,556 companies, with about ½ of those residing in the US. Exadata considers similar appliances to be NetApp Storage Systems, HP Storage Systems, and Dell EqualLogic Networked.

Every single hardware feature in Exadata is designed specifically to run a single application: Oracle. If you use an Exadata machine to run anything other than Oracle software, it won’t work because the hardware is designed for the software. SmartScan, FlashCache, IORM, and HCC are all designed to run only with Oracle database.

The second problem has to do with using any appliance (box) either on-premises or remotely hosted is that the customer must guess what their capacity requirements will be like in advance. Customers have to decide on compute, storage, and memory capacity first then Oracle ships the box to you. If the customer’s capacity estimate is too high, they constantly pay for idle expensive capacity. Too low, you have outages and lost revenue. If the customer needs more capacity, they place an order and the machine needs to be manually upgraded by Oracle or a hosting partner. The business agility with Exadata is very low. We will come back to this exact topic when we review Oracle’s claims regarding Exadata’s “elasticity.”

A Purpose Built Database Appliance

Exadata is described by Oracle as follows:

“The technology that enables Exadata’s unparalleled performance without any of the bottlenecks of traditional storage arrays is Exadata System software. This software powers the Exadata Storage Servers, providing an extremely efficient database optimized storage infrastructure.

One of the many unique features of Exadata System software is Smart Scan technology, which offloads data-intensive SQL operations from the database servers directly into the storage servers. By pushing SQL processing to the storage servers,
data filtering and processing occur immediately and in parallel across all storage servers, as data is read from disk and flash. Only the rows and columns that are directly relevant to a query are sent to the database servers. This greatly accelerates analytic queries, eliminates bottlenecks, and significantly reduces the CPU usage of the database servers.”

This is further explained in detail in the following quotation.

“iDB implements a function shipping architecture in addition to the traditional data block shipping provided by the database. iDB is used to ship SQL operations down to the Exadata cells for execution and to return query result sets to the database kernel. Instead of returning database blocks, Exadata cells return only the rows and columns that satisfy the SQL query. Like existing I/O protocols, iDB can also directly read and write ranges of bytes to and from disk so when offload processing is not possible Exadata operates like a traditional storage device for the Oracle Database. But when feasible, the intelligence in the database kernel enables, for example, table scans to be passed down to execute on the Exadata Storage Server so only requested data is returned to the database server.

The Oracle Database and Exadata server cooperatively execute various SQL statements. Moving SQL processing off the database server frees server CPU cycles and eliminates a massive amount of bandwidth consumption which is then available to better service other requests. SQL operations run faster, and more of them can run concurrently because of less contention for the I/O bandwidth. We will now look at the various SQL operations that benefit from the use of Exadata.”

How often in the history of Oracle have customers said the statement “Using Oracle was how we saved money.” Very infrequently we imagine.

Exadata as a Lower Cost Option?

Oracle’s claims regarding Exadata being a low-cost option is striking.

Wikipedia describes Exadata as follows:

“The Oracle Exadata Database Machine (Exadata[1]) is a computing platform that is specialized and optimized for running Oracle Database. The goal of Exadata[2] is to achieve higher performance and availability at lower cost(emphasis added) by moving database algorithms and intelligence into storage and networking, bypassing the traditional processing layers.[3]”

This entry has clearly been primarily written by Oracle or Oracle resources.

Right out of the gate, we run into an inconsistency. Oracle’s databases, hardware and applications along with SAP set the high water mark for initial cost and TCO. Furthermore, Exadata is a premium appliance that is well known to be extremely costly. Exadata is costly in terms of its initial price and maintenance costs. When Oracle sells a $9 million Exadata machine, $6 million of that is the Oracle database and other software. They also make 22% on software M&S and 20% on hardware and storage support and that’s standard support. The premier support level goes up to 35%.

This high cost is emphasized by the following quotation from Justin Bennett:

“Exadata is way too expensive and specialized to use for generic compute workloads. Exadata is an Oracle DB specific appliance available in a range of sizes to handle very low latency, complex data processing loads. Using that for VMs, PaaS applications would be a massive waste of resource and make Oracle cloud beyond the commercial reach of anyone. Also re Exadata, as with IBM and others, the list price is never remotely the purchase price.”

And this quote from The Exadata Certification Guide for 1Z0—027

“I am aware that Exadata is not completely public, which makes this certification a unique challenge. These Exadata machines are a huge money maker for Oracle, and they do not want to release anything that could allow competition to cut into their profit margins.”

A High Maintenance Solution

Furthermore, Exadata has high long-term operating expenses in addition to Oracle’s support costs. This is because Exadata is known to be so high in maintenance. Customers report a large number of issues years after Exadata is implemented. For example, Exadata requires its own training thread for Oracle DBAs.

Exadata is a significant investment for companies that are running the Oracle database on generic servers. Exadata is a complicated migration from on-premises servers. Exadata requires an implementation and migration that is considered complex. But this known complexity of Exadata does not stop Oracle for claiming the opposite.

“Exadata Cloud Service combines the world’s #1 database – Oracle, and the most powerful database platform – Exadata, with all the simplicity and cost-effectiveness of the public cloud.”

Oracle also claims that Exadata has zero CAPEX.

“With pay-as you-grow dedicated Exadata configurations, and infrastructure managed by Oracle experts, Exadata Cloud Service enables business agility and operational flexibility with zero CapEx.”

Reducing Costs for All Oracle Environments

However, Oracle’s argument is that Exadata is so efficient that it does reduce costs even if the Exadata purchase and maintenance is expensive. This of course also leaves out any cost savings that could potentially occur from migrating some of a customer’s databases away from Oracle. And currently, many customers are doing just this migrating Oracle databases to open source options on AWS RDS and Cloud SQL. But with Oracle’s cost proposals, all cost savings are for Oracle.

This proposed efficiency is explained in the following quotation.

“Because of the extreme performance, high storage capacity, and unique compression capabilities delivered by the Exadata Database Machine, workloads that would require very large traditional hardware systems can be run on much smaller Exadata systems.
The hardware needed for an application deployed on an Exadata system is often reduced 2-4X compared to a traditional system.

Exadata provides a huge RAM, Flash, and disk footprint for large data sets. Raw disk storage on an Exadata full rack can exceed 2.1 Petabytes while raw Flash storage can be up to 920 TB. In addition, Hybrid Columnar Compression often expands storage and memory capacity 10X. By intelligently moving active data across disk, Flash, and memory tiers, Exadata simultaneously delivers the highest performance and the lowest cost.

Exadata has the unique ability to consolidate many databases supporting multiple workloads in a single cloud platform. High-end OLTP, analytics, batch, reporting, and backups can all run simultaneously within and across databases with extreme performance. The extreme performance and capacity of Exadata enables very large numbers of databases and workloads to be consolidated on Exadata. Consolidating databases on Exadata reduces system hardware cost, software cost, and greatly reduces ongoing operations cost.”

Exadata as Cloud?

Oracle makes the following claims about Exadata as cloud.

“Exadata integrates Oracle Corporation hardware and software, including scale-out compute servers, scale-out storage servers, InfiniBand networking and specialized Exadata Software[4], packaged in one or more hardware racks, in a wide range of shapes and price points. Exadata compute servers use Intel Xeon processors and the Oracle Linux operating system or Oracle Virtual Machine to run Oracle Database software. Exadata Storage Servers use Intel Xeon processors to perform block storage functions and run Exadata Software that offloads data-intensive database processing. State-of-the-art Flash technology in Exadata Storage Servers has exposed the importance of Exadata’s offloading for realizing the full performance of Flash[5].

Exadata debuted in 2008 as the first in Oracle Corporation’s family of Engineered Systems[6] for use in corporate data centers deployed as “private clouds”. In October 2015, Exadata became available in the Oracle Cloud as a subscription service, known as the Exadata Cloud Service[7][8]. Oracle databases deployed in the Exadata Cloud Service are 100% compatible with databases deployed on Exadata on-premises, which enables customers to transition to the Oracle Cloud with no application changes. Oracle Corporation manages this service, including hardware, network, Linux software and Exadata software, while customers have complete ownership of their databases.”

The part about being cloud is false. In order to qualify as cloud, multitenancy is a primary requirement. And Exadata is not multitenant.

Running Multiple Database Processing Types

“Historically, specialized database computing platforms were designed for a particular workload, such as Data Warehousing, and poor or unusable for other workloads, such as OLTP. Exadata has optimizations for all database workloads, implemented such that mixed workloads share system resources fairly. Resource management features also allow for prioritized allocation of system resources, such as always favoring workloads servicing interactive users over reporting and batch, even if they are accessing the same data.“

The issue brought up in this quotation is not a problem on AWS/GCP/Azure, because the hardware is customized for the workload. Exadata is not optimized for every database workload, because no hardware can be. Therefore, this another inaccuracy by Oracle made in explaining Exadata.

Secondly, Exadata is only designed to run the Oracle database (11, 12, 18), which is not the best database for all workloads. Increasingly, the multi-database approach or “multi-base” approach is being leveraged by companies that use AWS and Google Cloud. Oracle has less interest in the multi-base as they do not offer very many different database types. Werner Vogels of AWS describes the multi-base approach.

Werner Vogels on Leveraging Multiple Database Types

While SAP has proposed using a single database type (the one database type SAP wants customers to focus on, which is HANA), and finds its most willing customers in those that know the least about databases, AWS offers their customers the choice of leveraging many different database types.

In an excellent article by Werner Vogels who is the CTO of AWS explains this.

“A common question that I get is why do we offer so many database products? The answer for me is simple: Developers want their applications to be well architected and scale effectively. To do this, they need to be able to use multiple databases and data models within the same application.”

Notice the last part of this paragraph, where Werner Vogels describes using..

“multiple databases and data models within the same application.”

Wait, let us back up. What was that?

We all know that applications have a single database right? How does a single application use multiple databases?

What is Werner talking about?

Well, it turns out Werner is describing software development that is different than the monolithic environment for which the majority of those reading this book most likely are most familiar with.

Werner goes on to say..

“Developers are now building highly distributed applications using a multitude of purpose-built databases.”

That is the application that we think of is one way of developing, but this is giving way to distributed applications that can access multiple databases.

Multibase Versus the Monolithic Model

It is an unusual way of thinking about applications. Particularly for those of us who came up under the monolithic model (which is most of the people that work in IT). It means rethinking much of what has been part of the standard model in IT for many years. The same debate currently occurring between monolithic and microservices development occurred in operating system development previously. The following quotation does a good job of explaining how this argument goes.

“As people designed more and more operating systems, they quickly realized that there was one tough question: how big should it be? Some people argued that the OS should be as big as possible and come complete with all of the features that someone might want to use. Other countered with stripped down designs that came with a small core of the OS surrounded by thousands of little programs that did the same thing. To some extent the debate is more about semantics than reality. A user wants the computer to be able to list the different files stored in one directory. It doesn’t matter if the question is answered by a big operating system that handles everything or a little operating system that uses a program to find the answer. The job still needs to be done, and many of the instructions are the same. It’s just a question of whether the instructions are labeled the “operating system” or an ancillary program.

But the debate is also one about design. Programmers, teachers and the Lego company all love to believe that any problem can be solved by breaking it down into small parts that can be assembled to create the whole. Every programmer wants to turn the design of an operating system into thousands of little problems that can be solved individually. This dream usually lasts until someone begins to assemble the parts and discovers that they don’t work together as perfectly as they should.”

Furthermore, software that often looks monolithic, isn’t, as the following quotation explains quite well.

“The marketplace has developed ways to sell software as big chunks to people, but these are fictions that camouflage software integration. In modern practice, programmers don’t just create one easily distinguished chunk of software known as Microsoft Word or Adobe Photoshop. They build up a variety of smaller chunks known as libraries that link these together. Microsoft Windows, in fact, includes a large collection of libraries for creating the menus, forms, click boxes, and what not that make the graphical user interfaces. Programmers don’t need to write their own instructions for drawing these on the screen and interacting with them. This saves plenty of time and practice for the programmers, and is a large part of what Microsoft is selling when it sells someone a box with Windows on it.”

The Limitations of the Relational Database

Werner goes on to describe the limitations of the relational database.

“For decades because the only database choice was a relational database, no matter the shape or function of the data in the application, the data was modeled as relational. Is a relational database purpose-built for a denormalized schema and to enforce referential integrity in the database? Absolutely, but the key point here is that not all application data models or use cases match the relational model.”

This model was in great part pushed by vendors like Oracle and IBM as the relational database was what they had to sell. If you read Oracle literature, for instance, the relational database is the best database design for everything, and the best relational database is (or course) Oracle. And furthermore, the best value in all of the database ecosystem is…..Oracle.

We have seen in the rapid growth of databases like MongoDB and Hadoop that specialize in either unstructured data or data with lower levels of normalization. However, this brings up the question of why these different database types have begun to proliferate.

The proliferation is in part covered by Werner when he describes how Amazon ran into the limitations of using the relational database.

“We found that about 70 percent of our operations were key-value lookups, where only a primary key was used, and a single row would be returned. With no need for referential integrity and transactions, we realized these access patterns could be better served by a different type of database (emphasis added). This ultimately led to DynamoDB, a nonrelational database service built to scale out beyond the limits of relational databases.”

Let us consider that AWS has a very fast growing relational database service in RDS. However, they also have fast-growing non-relational databases like DynamoDB, as well as many other database types. AWS is offering a menu of databases, each targeted towards specific applications. So how many database types are there? Werner has an answer for this as well.

The Different Database Types According to Werner

Below we have provided a synopsis of the different database types, their intended usage, and the database that reflects them by Werner.

  • Relational: Web and Mobile Applications, Enterprise Applications, Online Gaming (e.g., MySQL)
  • Key Value: Gaming, Ad Tech, IoT (DynamoDB)
  • Document: When data is to be presented as a JSON document (MongoDB)
  • Graph: For applications that work with highly connected datasets (Amazon Neptune)
  • In Memory: Financial Services, Ecommerce, Web, Mobile Applications (Elasticache)
  • Search: Real-time visualizations and analytics generated by indexing, aggregating, and searching semi-structured logs and metrics. (Elastisearch Service)

Werner sees applications leveraging multiple database types. But SAP and Oracle do not see it that way. SAP and Oracle want customers to use a single database type, a “supercharged” RDBMS database that they state is best for all database processing types. And as pointed out by AWS’s documentation, databases are often selected based upon factors that are not part of a thorough examination of the requirements.

Therefore the argument for Exadata extends Oracle’s false argument about the universality of the Oracle RDBMS database to the appliance (hardware) with Exadata.

These universal application claims are repeated in this slide from Oracle.

Overall, this is an oversimplification designed to sell to executives who do not understand the various database processing types and how different they are from one another. Since 2015, Microsoft has been the leader in database sales by revenue according to Gartner. Oracle is number two since 2015 and AWS beat both IBM and SAP for third place. So selling a single product no longer works because they get beat on capability and price per individual products.

Exadata as Elastic?

Secondly, Exadata does not provide anything like the elasticity of the cloud. When one buys on-premises Exadata machines, one acquires the entire thing. Although one can buy smaller or larger Exadata configurations as described in the following quotation.

“Just as with Oracle Exadata, customers of the Oracle Database Cloud Exadata Service can start relatively small then expand. They can start with a quarter rack of the system, with 28 microprocessor cores enabled and 42 terabytes of storage, capable of 1 million input/output operations per second (IOPs), then work their way up to potentially hundreds of cores and hundreds of terabytes of storage, capable of multiple millions of IOPs.”

When Exadata is “cloud” or hosted, again, it is not elastic.

“Long running requests, characterized by Data Warehouses, reports, batch jobs and Analytics, are reputed to run many times faster compared to a conventional, non-Exadata database server.[11][12] Customer references often cite performance gains of 10x or greater.”

Exadata for Complex and Even Poorly Written Query Optimization

This all depends upon what hardware configuration was used to begin with. Exadata is very good at running queries.

“The initial 2008 release of Exadata (V1) was a joint development between Oracle (software) and Hewlett-Packard (hardware). The second generation (V2) of Exadata switched to hardware from Sun Microsystems, and shortly thereafter Oracle acquired Sun Microsystems and thus gained ownership of the main hardware components of Exadata.”

This was done to leverage Oracle’s Sun acquisition.

“Ellison had wanted a hardware/software integrated stack for a long time and this gave him what he wanted. Now his Exadata and Exalogic servers would be produced by Sun and not partners HP and Fujitsu. Even better, he could offer a turnkey system with Oracle software pre-installed and fully integrated. Oracle has focused on using its appliances and engineered systems as turnkey products where the OS, database, middleware, applications, Java and everything else the customer needs or wants is pre-installed and fully integrated. Just plug and go. That smacks of the mainframe days but in some cases that’s exactly what some people want.”

Oracle also proposes that Exadata is perfectly suited for the cloud.

Here Oracle claims Exadata is multitenant. However, Exadata can only dedicate OCPUs and a single PDB (Pluggable Database) to each cloud client but the memory, flash cache, and other resources on an Exadata system are not dedicated. What this means for customers is that there’s absolutely no guarantee of consistent performance because Oracle will decide who and how many other clients share the PDB resources available within a single Exadata machine otherwise known as the noisy neighbor effect.

Oracle considers Exadata as perfect for supporting database as a service.

“The Exadata Database Machine can host many databases, enabling massive database consolidation or a sophisticated Database as a Service private cloud. Multi-database environments inherently have diverse, complex, and unpredictable workloads mixing OLTP, analytics, and batch operations with sequential and random access patterns. Exadata’s ability to run any type or mix of database workloads with industry leading scalability and performance makes it an ideal consolidation platform – whether for multi-database workloads, or for pluggable databases with Oracle Multitenant in Oracle

Database 12c and Oracle Database 18c Multi-database environments create an inherent risk that one database will consume too many resources and therefore impact the quality of service of other databases. The Exadata Database Machine uniquely provides end-to-end prioritization from the application to database CPUs, network, and storage. Priorities and resource limits can be specified at the physical database, pluggable database, connection, application, user, or even job level to ensure that each of the consolidated databases or SQL operations receives the necessary resources and achieves the target response times.”

Here Oracle proposes deployment options which include public cloud. What is “clout at customer?” That is normally called on-premises.

Oracle has a constant issue in its marketing of Exadata and Exalogic of stretching and actually altering the definition of cloud. Oracle actually has very little cloud to offer, and therefore it follows a central strategy of pretending its offering are cloud when they aren’t. This is also expressed in the following quotation from A Technical Overview of the Oracle Exadata Database Machine and Exadata Storage Server.

“The Oracle Exadata Database Machine is an easy to deploy solution for hosting the Oracle Database that delivers the highest levels of database performance available. The Exadata Database Machine is a “cloud in a box” composed of database servers, Oracle Exadata Storage Servers, an InfiniBand fabric for storage networking and all the other components required to host an Oracle Database.”

Oracle’s Cloud in a Box?

The cloud is not in any box. It is a remote location which is nearly infinitely scalable at the hypercloud service providers (and infinitely scalable downward and VMs can be paused or shut down). Being able to log in remotely to a server that is managed by one customer is called an on-premises server. Oracle would counter this argument with this proposal.

“While an Exadata Database Machine is an extremely powerful system, a building-block approach is used that allows Exadata Database Machines to scale to almost any size. Multiple Database Machine X2-2 systems can be connected using the InfiniBand fabric in the system to form a larger single system image configuration. Multiple Exadata Database Machine X2-8 racks can similarly be connected. This capability is done by connecting InfiniBand cables between the racks as all the InfiniBand infrastructure (switches and port cabling) is designed to provide this growth option. Up to 8 racks can be connected by simply connecting the InfiniBand cables. Larger configurations can be built with additional InfiniBand switches. Any combination of X2-2 Full and Half Racks can be connected. Quarter racks can be inter-connected with other racks in two circumstances. Two Quarter Racks can be inter-connected to each other or one Quarter Rack can be connected to any combination of Full and Half Racks. The inherent capability of the Exadata Database Machine to grow enables the support of the largest databases any application would require.”

However, this is once again, not very different from on-premises servers. On-premises servers can also be scaled upward. In fact, Exadata is an on-premises server, except it is a high-end item and an engineered appliance designed for a specific type of processing.

When AWS, for instance, offers the RDS managed database service, they are gaining enormous scale economies by allowing so many customers to use a database instance while protected by VPC. A significant advantage of RDS is that it is multitenant. Multitenancy functionality is quite advanced and allows many user’s data to be kept in the same RDS. This is how the RDS gains such high degrees of economies of scale, and how it reduces maintenance overhead. This is why the cloud is so much more than hosting or “private cloud.” It means creating large-scale economies from managing many customers on one instance of software. In this case a database.

RDS has a default security. However, unless one is using a legacy database, the RDS service is protected with a VPC or virtual private cloud is used to segment the customer’s DB instance from other DB instances, even though the customers share or are multitenant on the RDS.

The VPC creates a virtual network within AWS for each customer. The metaphor of a “network” applies because it is as if it is a network in one’s own on-premises data center. In the graphic above, containers are shown within either a datacenter or a VPC, implying that the two are logically the same.

The VPC provides a high degree of control related to things like selecting IP address ranges, configuring routing, and other security features. As each database instance has a flexible amount of compute resources allocated to it, with so many customers in the same VPC protected database, the administration overhead declines significantly.

The combination of lower overhead database software combined with multitenancy and central administration takes RDS into an area where Oracle cannot compete. This is because Oracle is not offering what AWS is offering. Oracle has a cloud DaaS service, but it is nothing at all like AWS or Google Cloud’s offering.

If customers chose to perform a homogeneous migration from their Oracle on-premises database to the RDS (or bring up a new Oracle RDS), then that customer still gets two of the three benefits listed above.

AWS RDS and Cloud SQL Popularity

The relational database management service or RDS and Cloud SQL offering are one of AWS’s and Google Cloud’s most popular services. Through this managed service, AWS and Google can concentrate database resources to manage enormous numbers of customer’s data centrally. In a new development in the history of databases. In either the on-premises model or in the hosted model (say with an IBM or CSC data center) each customer installs data into a database that they purchase that only contains their data. Then multiple database administrators employed at the customer manage those single customer database instances. The fundamental inefficiency of single-tenancy is a primary root of the inefficiency of the on-premises database. AWS RDS and Google Cloud SQL are not merely offering that same design but hosted with them. Instead, RDS and Cloud SQL are massively multitenant. This means that each customer has their data in one database. However, the data can be protected by a VPC, but it still (through some amazing technological developments that were previously unthinkable) resides in one database.

Therefore, while it seems like the customer data is all together in the database, the VPC per customer means it is isolated. This is why Oracle and SAP’s pronouncements about “private cloud” are silly and misleading. Private cloud is Oracle and SAP’s way of cloud washing what they are offering, which is hosting. This is just a method of moving a server from on-premises to the vendor’s premises. It’s not cloud. One can obtain privacy from an efficiency public cloud by using a VPC.

This design provides large economies of scale in database management. Also, it means that Oracle cannot compete with AWS or Google Cloud on both price and functionality. Oracle offers a sophisticated database, but their ability to manage their database in a cloud environment is vastly inferior to AWS or to Google Cloud. This level of detail is rarely explored in public articles that cover this area. The distinctions in this area are generally lost on all but those that work in a hands-on capacity in the area.

Every conversation with Oracle seems to come back to its database. Whenever buying any software, Oracle’s official position seems to be that one should begin by acknowledging that Oracle has the best database for all purposes, and then all applications should then be purchased from Oracle, regardless of the capability of the application. This also extends to the cloud. Cloud purchases should be based upon who makes the Oracle database, which happens to be Oracle, and that should drive companies to use Oracle Cloud. This pattern from Oracle is expressed as follows by Denis Myagkov.

“I think that SAP’s and Oracle’s myth department is propelling they database solutions without any context. It’s pretty weird to compare one database with another and not mention of its application. Any database is only a way to store some data somewhere and somehow and here we have the huge gap – what system will be consumer of they databases?”

Pivoting the Discussion and Emphasis to the Database

Oracle prefers if the entire discussion around infrastructure to be focused on the database rather than viewing the database as part of an overall infrastructure capability. And open source or databases that are closed source but accessible with little license overhead (like those offered by AWS and Google Cloud) are part of what allows for infrastructure flexibility.

Increasingly in the AWS and Google Cloud world, Oracle licensing and auditing is a liability.

“Oracle Exadata was architected for the cloud since its inception. That’s because the system is engineered and sold in building-block increments of hardware servers and storage servers. Customers can add and expand servers as their demands for performance and capacity increase.”

After Exadata servers are installed, Oracle’s auditing services will very much look forward to auditing the usage of Exadata and the Oracle database. Auditing is not an issue if Exadata nor Oracle databases are not purchased. Neither PostgreSQL nor MariaDB audit customers. This gets back to a theme discussed throughout the book that when software vendors begin to achieve monopoly power, they begin to ask to be paid multiple times of the same IP. Oracle and Oracle consulting partners specifically setup Oracle products to have an audit liability and Oracle has a timing cadence which occurs after the customer attains maximum lock-in where they audit an account, and they more or less know what they will find.

“It’s scale-out compute, scale-out storage, with a superfast network, so it had a cloud architecture from day one,” Loaiza says. “We’ve deployed it as a kind of on-site cloud for customers, and now we’re deploying it in the public cloud.”

And Oracle proposes they have many Exadata machines placed in various regions.

“Oracle Cloud Infrastructure Services are deployed in Regions and Availability Domains. A region is a localized geographic area, and an Availability Domain is one or more data centers located within a region. Availability Domains are isolated from each other, fault tolerant, and very unlikely to fail simultaneously. All the Availability Domains in a region are connected to each other by a low latency, high bandwidth network. This makes it possible to deploy a production Exadata Cloud Service instance in one Availability Domain, along with a synchronized Exadata Cloud Service instance in another Availability Domain, providing zero-data-loss disaster recovery enabled by Oracle Active Data Guard. Active Data Guard can also be used to enable an efficient WAN-based Disaster Recovery configuration by setting up synchronized Exadata Cloud Service instances between Regions. Oracle Cloud Infrastructure Services network infrastructure is comprised of a high performance, non-oversubscribed, flat physical network which limits latency between Availability Domains to sub-milliseconds. Latency is even less between compute instances within an Availability Domain. In addition, Oracle Cloud Infrastructure Services enable a secure, private software-defined Virtual Cloud Network (VCN) that allows customers to treat Oracle Cloud Infrastructure Services as a secure and elastic extension of their on-premises network. Customers can configure their Exadata Cloud Service instance in their preferred VCN with complete flexibility that includes assigning their own private IP address space, creating subnets, creating route tables and configuring stateful firewalls. Customers can configure the VCN with an optional Internet Gateway to handle public traffic, and an optional IPSec VPN connection to securely extend their on-premises network”

As of this publication, Oracle only had four regions, and Oracle is investing comparatively little in its cloud infrastructure.

Private Cloud Versus Public Cloud

In Oracle’s case, the appliance model (Exadata, Exalogic, etc..) is their primary go to market and the most profitable line of business. Every hardware and storage sale makes money on the hardware plus 20% of the contract value annually on support (10% on servers + 10% on storage for standard support). For these two, customers running software on other clouds means the end of their hardware and storage businesses along with the highly profitable M&S revenue.

This is Oracle and IBM’s “cloud” strategy. This is why they push “private” cloud and “cloud at customer” and double the license of Oracle software on AWS and Azure. This is also the reason Oracle cloud runs exclusively on Exadata bare metal machines and not industry standard Intel platform. SAP has no problem running software on other clouds because that does not cannibalize any profitable revenue streams.

The Problem: A Lack of Fact-Checking of Oracle

There are two fundamental problems around Oracle Exadata. The first is the exaggeration of Exadata, which means that companies that purchased Exadata end up getting far less than they were promised. The second is that the Oracle consulting companies simply repeat whatever Oracle says. This means that on virtually all accounts there is no independent entity that can contradict statements by Oracle.

Being Part of the Solution: What to Do About Oracle

We can provide feedback from multiple Exadata accounts that provide realistic information around Exadata and other Oracle products — and this reduces the dependence on biased entities like Oracle and all of the large Oracle consulting firms that parrot what Oracle says. We offer fact-checking services that are entirely research-based and that can stop inaccurate information dead in its tracks. Oracle and the consulting firms rely on providing information without any fact-checking entity to contradict the information they provide. This is how companies end up paying for items which is exorbitantly priced, exorbitantly expensive to implement and exorbitantly expensive to maintain.

The Necessity of Fact Checking

We ask a question that anyone working in enterprise software should ask.

Should decisions be made based on sales information from 100% financially biased parties like consulting firms, IT analysts, and vendors to companies that do not specialize in fact-checking?

If the answer is “No,” then perhaps there should be a change to the present approach to IT decision making.

In a market where inaccurate information is commonplace, our conclusion from our research is that software project problems and failures correlate to a lack of fact checking of the claims made by vendors and consulting firms. If you are worried that you don’t have the real story from your current sources, we offer the solution.

Financial Disclosure

Financial Bias Disclosure

Neither this article nor any other article on the Brightwork website is paid for by a software vendor, including Oracle, SAP or their competitors. As part of our commitment to publishing independent, unbiased research; no paid media placements, commissions or incentives of any nature are allowed.

Search Our Cloud Services Content


The Public Cloud Revolution Book

The Public Cloud Revolution: How Open Source is Displacing IT Mega Vendors

Interested in how to open source is powering public cloud providers like AWS and Google Cloud and what this means for the different modalities of computer hardware (cloud, on-premises proprietary server, mainframe, and appliances?). This book covers many topics that are greatly underrepresented in the common IT media coverage of these topics.