It's a trick question ;0). Exadata main power comes from its software integration between the database and the storage servers - primarily by reducing the data scanned and sent, and to a lower degree from the hardware (the number of spindles, SSD, cache and IB network). The specialized software components (smart scan, storage indexes, HCC, etc.) could be used in other architectures, such as with EMC storage, however as a marketing tool Larry put a lock on their availability. Other software components are still available to all (ASM, Smart Flash Cache, etc). Releasing an appliance helps reduce the amount of unknowns in a system, and potentially time-to-deploy. Remember that Exadata came originally as a way to fight Netezza and the like that were able to ship a box and have it running the next day (if not the same). Oracle needed a tool to compete with both the BI performance and the delivery time that the competition (startups/appliances) were able to conjure. Comes Exadata. Obviously it comes with many problems, some of them tie to the way Oracle as an RDBMS function (shared everything type architecture) and ASM, management, BC/DR, quallity and business tie gaps as well ('black box to everyone but the DBA's). I guess they like it since they continued to produce Exalogic and Exalytics - both with less focus on software specialization and are more simply a packaging play with fancy marketing names.
So to summarize, I don't think that it matters whether Oracle is a good fit for appliances (answer should be no I'd think, it's not an iPhone ), but rather that they needed to bridge a gap they identified and they did. Whether the fix is good or bad we can debate. I do believe that if they were to work collaboratively with partners like EMC the end result would have been much better (for many reasons, including the lock down, technology limitations and many other aspects), but so far Larry looks at the appliance market as a cache cow (why share technology with partners?). I wonder how long Oracle OpenWorld would continue to be called 'open'.. hehe.
Excellent answer and I believe you hit upon the pivotal point for Oracle Exadata: Priority vs. Open integration. Oracle choose a priority approach with Exadata. The long term implications of a priority approach are a shorter self life as new technology tends to eclipse priority stuff, the inducement of competing technologies and finally competition in the form of other converged systems. I do believe, in part, Oracle’s success with Exadata is its marketing it as an appliance. Hence the original question “Is Oracle truly a good candidate for appliances?” Take the iPad for example, it’s easy to use, easy to manage and makes many tasks much more portable despite its lack for Adobe Flash. Larry has connected the dots from the iPad through to Oracle in explaining the advantages of the appliance approach. The iPad is extremely successful because of its elegant approach to computing. Will Oracle Exadata experience the same success? To turn the question around, “Is the only way for Oracle to compete with other DW and BI solutions is to create a priority stack?”
Oracle start its Appliance journey with Exadata on BW, I believe Oracle selected an approach with less resistance inherited from existing IT system, BW system does not request precise and complex system as Pro (OLTP) system does, such as DR, Backup/Restore, etc.
But just as SAP HANA, which is still only available for BW. I think the Appliance still have a long long way to go, because SMP/Shared Architecture is still on the dominant position.
I would argue that if we're really talking about Oracle databases on appliances, we shouldn't really be talking about Exadata, but instead focusing on Oracle's new Oracle Database Appliance (ODA)<-- see, it even has appliance in the name . As with any appliance (versus a custom solution), you design the appliance to fit the needs a certain market segment with certain needs. Need something from Oracle that can handle really high I/O for an OLTP or DW type environment? Exadata might fit the bill. Need something from Oracle with more of a streamlined setup and maintenance but that is limited in terms of expandability and performance ? ODA might bear investigation.
Appliances are useful for solving a very focused need. It's me belief that there very very very few situations with respect to Oracle databases that really need the performance you can get from an Exadata type machine. For example, EMC runs the 4th or 5th largest Oracle EBS environment in the world - no Exadata. Oracle runs the largest Oracle EBS environment in the world - as of two months ago and maybe still, no Exadata.
What you trade for performance is flexibility - There are places where the performance of something like an Exadata or really any appliance make sense - but you need to consider ALL your requirements, and with Oracle databases that usually have associated web servers, application servers, etc, there's more to consider. Exadata is very limited in regards to backup and DR solutions. Need to run some sort of application server so that users can actually access that data stored in the Exadata? If so, that application server needs to run on something else - maybe Exalogic, maybe not.
I haven't yet dealt with a client where the performance requirements were something that required Exadata level equipment, so maybe I just have my eyes wide shut.
I have to laugh on this one. Since I really grew up in a truly appliance-oriented environment (NetApp), I think I understand what an appliance is pretty well.
At NetApp (and filers in the days I was there were true appliances by any measure), the appliance concept meant that the device was a toaster: One lever to push down, and one dial to turn. That's it. Plug it in. It works. No step 2.
Cisco really originated the term with the original Cisco router, which was also a true appliance.
An appliance has the following characteristics:
- Extremely simple interface. Should be vastly simpler than doing it the non-appliance way. I.e. a Cisco router is vastly simpler than running routed on a UNIX box.
- A single purpose. The device must be dedicated to doing one thing, but doing it extremely well. Like the way a Cisco router is much better at doing routing than a UNIX box running routed. Or like the way a NetApp filer is much better at doing NFS file serving than a UNIX box running nfsd. You get the idea. By dramatically reducing the number of functions the device performs, you also dramatically reduce the amount of code that must be run on the device. (The original ONTAP OS was a single-threaded 16 bit OS with only a few 100K of lines of code.) This leads to the next feature of an appliance which is:
- Vastly reduced cost. The original NetApp filer was about a $5,000 device. An equivalent Sun box used as a filer ran around $50,000. Similar cost differences existed for Cisco routers vs. UNIX boxes as routers.
- Transformative technology. An appliance, if it is truly an appliance, becomes the obvious and natural way to do things. Within a very short period of time after introducing the router, Cisco controlled the market. They completely displaced the previous way of doing routing. The same thing occurred in file serving with NetApp.
By any reasonable measure, Oracle ExaData fails all of these tests:
- It has as complex an interface as any Oracle database server (which is to say it runs the most complex and expensive piece of software ever written for general purpose use).
- Oracle ExaData contains general purpose servers, which can be used to run basically anything you want. You can load any Oracle application on it certainly, and no-one would claim that an Oracle database server is an appliance!
- Oracle ExaData is manifestly more expensive than a normal, open-systems database server, and vastly more expensive (assuming intelligent management) than using VMware vSphere for virtualizing Oracle database servers.
- Oracle ExaData is possibly addictive in the Big Blue sense, but it is certainly not a transformative technology in the same way that a Cisco router, iPad, iPhone, or such is.
In terms of an analogy that works, I like to use cars. The two companies in the car business that manufacture applicance cars are Honda and Toyota. The Honda Civic is an appliance car, as is the Toyota Camry. Either one of these cars provides all of the appliance advantages:
- They have a radically simplified interface. Everything about these cars is designed to make it effortless to operate. Because it is so simple, is is also very reliable and efficient.
- They are single purpose vehicles. They get you from point a to point b, that's it. Nothing fancy.
- They are sold at a very reasonable cost, relative to non-appliance vehicles (such as BMW, or Mercedes for example).
- Once you have driven a Honda Civic or Toyota Camry, assuming you are an appliance driver (and there are many of these), these cars are completely addictive. You simply trade one in for the new model once the old one wears out (and they take a long, long time to wear out). I have known folks who have been driving these cars (in various model years) their entire lives.
Using the car analogy, ExaData is definitely not a Honda or a Toyota. It is not even a BMW or a Mercedes. It is a Ferrari. It is a tricked out, high performance machine. It is very fast, no question. It is *&^% expensive though. And it is very, very complex and demanding to drive.
This is a very intriguing conversation. Most of our customer base buy their equipment in a modular approach and put together the infrastructure themselves. In addition, they support the infrastructure in the same modular support model. Magnifying the problem, these teams have different objectives: DBAs want to present the data as fast as possible and the storage admin want to give out as little as storage as possible.
As systems grow or change, these teams are adding new variables to the infrastructure creating systems that are not balanced. CPU, Memory, Storage and network all have their limitations and need to be aligned. If all the products are inbalanced, it creates a bottleneck with wasted resources that often cannot be found by the modular support model. This problem is compounded with bad database schema design, bad query writing and a common problem of not adjusting the init.ora parameters as they upgrade to a new version of Oracle.
Oracle's was targeting these specific customers that had performance problems with the hardware running the Oracle Database. Do not under estimate Exadata'a ability to mesmerize the customer to the point that they are oblivious of the other Exadata features: No reliability, difficult support model, introduces new challenges in the data center.....
We need to take a look at the whole picture ....from Wiki's Computer Appliance page ..."A computer appliance is generally a separate and discrete hardware device with integrated software (firmware), specifically designed to provide a specific computing resource. These devices became known as "appliances" because of their similarity to home appliances,".... According to Oracle's appliance philosophy: You can wash your jeans in a dishwasher, you can dry your plates in the clothes dryer and peel the potatoes in the garbage disposal. We need to give Kudos to Oracle marketing.
As the entire IT world is moving to a configuration of Cloud Computing, Oracle is going against the grain with an appliance based solution and calling it Cloud. Exadata came out in 2008 and as of 2010 Larry Ellision was still in denial of Cloud Computing. In 2011, everything is cloud based at Oracle. Exadata did not under go any major changes yet it transformed into a cloud database. Oracle's on Demand Software became Oracle SaaS.
Cloud computing requires to alter the support model. That new role encompasses the entire data center under any infrastructure tier. In Oracle's appliance world, each appliance need it own Appliance specific expert that is dedicated. For example, I can administer any Exadata box but do not look at me to help with the Exalogic or T4 Super Cluster appliance.
Oracle was critical of these modular sales model only to lift the modularity to another level and indicated it has resolved the issue but essentially just shifting it. The moral of the story is appliance are against the grain of cloud computing. The implicit message is not all customers are technically savvy.
If you want to see the success of Exadata please talk to someone at Solyndra or MF Global, two companies I helped sell Exadata.