Oracle Surprised by the Present
blogs.technet.com
blogs.technet.com
That being said, I'm always a bit surprised that SQL Server doesn't get the recognition that (IMO) it deserves. It is a pretty solid product. I was in a position to use it for years and saw it mature such that it eventually replaced legacy DB2/Adabas environments. It performed admirably and for a fraction of the price.
If the powers that be hadn't decided Oracle at all costs (and literally at all costs) against my counsel, they would have saved a lot of $$ and I might still be working there.
Your system/business setups simply reflect this, either through clustering or planned outages.
Not sure why I'm providing a rational opinion to a snarky jab at Mircosoft, but imo, they have made fantastic progress on the security front over the past decade.
I've heard Server 2012 adds more capabilities that would make it administerable more like a Linux sysadmin would expect.
Sorry if I offended the point-and-click "sysadmins".
And, I say this as someone who has literally nothing to do with them on a daily basis anymore.
Anecdotal, but that guy saved us a lot of time and effort because he knew what he was doing, and I'd take his real life experiences over any sales rep! I'd probably also enjoy the awesome SQL Server GUI in a *nix environment if I could have that. :)
I don't understand all the hate for Oracle. They sell products that work well and they make a lot of money doing it. What's the problem?
That's... debatable. The problem is that the Oracle portfolio is now so vast, after a decade of acquisitions targeted at "surrounding" SAP, that quality levels are very uneven across the board. Some products require constant attention from specialized personnel or they'll blow up in your face, the learning curve is always very steep, and some of them are just full of bugs. And I say that as one of those specialized folks.
Also, their licensing terms (and the salespeople who espouse and enforce them) are unabashedly, shamelessly greedy.
Windows NT OTOH makes even Vista look good. Not that it isn't solid, but it's like driving a tractor.
As someone who has spent 13 years with it and has dragged himself through multiple certification cycles, the reason it doesn't get recognition is that it's an insanely deep money pit when you have to scale it. By the time you've forked out for the (now suddenly per core rather than CPU[1]!) enterprise license, all the 3rd party backup and restore tools, numerous windows licenses and had your arse audited for license compliance, you're down a lot of cash.
Not only that, from a technical perspective, it's an absolute bastard to administer, particularly if something goes snap. It requires significantly more meat popsicles to run than a similar sized Postgres or MySQL cluster.
To be honest, it's cheaper to solve all the problems that SQL Server solves at an architectural level these days. It's no longer something special (was it ever?).
The only thing going for it these days is that it's better than Oracle.
[1] This properly fucks us on the 2008->2012 upgrade cycle as our primary machines are 4x 8 core Xeons. So that's from 4 licensable CPUs to 32 licensable CPUs. We have 4 of those machines. Bastards!
SQL Server licensing is pure madness.
I will never forget sitting across from a cadre of regional level Microsoft folks who had just finished explaining the intricacies of SQL virtualization licensing.
The seemingly straightforward design we envisioned to cost perhaps $120K under any sort of reasonable, logical licensing structure attached to per socket price of SQL would actually run upwards of $1.3M to properly comply with the Microsoft licensing terms.
It wasn't even the absurd total or the ridiculous terms that were so memorable. It was the fact the Microsoft folks seemed to be as shocked at our incredulity as we were at their terms.
I got the impression they weren't used to providing more than "It's cheaper than Oracle" as justification to receive millions of dollars.
We asked MS what sort of licences (& how many) we'd need to purchase and were still waiting for an answer months later when I left.
Screw them - we're moving some of our shit to Java/Linux (some of it is already there).
1] This properly fucks us on the 2008->2012 upgrade cycle as our primary machines are 4x 8 core Xeons. So that's from 4 licensable CPUs to 32 licensable CPUs. We have 4 of those machines. Bastards!
But the Core pricing is 1/4 the CPU pricing, so you're going from 4 licensable CPUs to the equivalent of 8 licensable CPUs.
That's only a quarter of a million pounds to half a million pounds change...
Really? If you have some pro DBAs in charge you don't need mass quantities of warm bodies. (The problem is that 90% of the folks that claim to be experienced can't tell a clustered index from a cluster f__k.) You can automate 95% of the job and spend the rest of the time helping users do better.
"To be honest, it's cheaper to solve all the problems that SQL Server solves at an architectural level these days. It's no longer something special (was it ever?)."
Bullshit. I'm dealing with clients that insist on writing custom applications to do complex ETL instead of using SSIS or decent third party tool. Or wasting months writing code that reinvents SQL that's been around for decades.
SSIS is a pile of shit. It's impossible to test properly, suffers horrific deployment problems, is very unreliable and the only way to get transactional guarantees is using DTC which literally destroys your performance.
A better solution is something like spring batch / spring integration back off a queue solution.
Agree that NIH/DIY is bad here.
I have some beefs with SSIS (which I bring up to the SSIS team lead, Matt Masson, every time we meet) but it's not as bad as so many folks seem to make it out to be. However, what I always see is a never-ending shortage of clients that write their own ETL systems that have 10-20% of the functionality, and 10-20% of the performance (and 10-20x the bugs). Even my current client seems to be doing just this.
Replacing homegrown ETL systems with SSIS, ODI or PowerCenter implementations is a great way to make money. I've seen ETL systems that folks were mighty proud of that had throughput (granted, to some very questionably modeled "data warehouses") measured in bytes/sec. Yes, bytes. We're talking 3 hours to get an 2MB file into the final fact table with some custom "framework" written in C# or Java that uses a web service for all message passing for servers within the same rack (and will always be in the same rack). Again, not necessarily the framework's fault either, just how poorly it's used, and how poor the data is modeled (in the above case, I was able to rewrite it with their tools and get the 3 hour job completing in ~50 seconds after a few days work and tearing through their framework's source code repo). The bugs you can find are fun too. My favorite was when the web service endpoints weren't reusing the same connection, so they were exhausting all the ephemeral TCP ports on the database server's TCP/IP stack -- when the job ran ~200 times faster, since the messaging was so chatty, the ETL framework was basically DoSing their SQL Server (for any service/application using a non-specific port).
Honestly, if you enjoy making things 100 - 10000x faster, live in the database world. The market will present an infinite number of opportunities for you to do so with commodity DBMSes such as SQL Server, MySQL, Postgres and Oracle. Most of it is simply cleaning up bad data modeling decisions, cleaning up a complete misunderstanding of how their database engine works, and overly complicated systems doing very simple things.
-----
I do agree that SQL Server licensing, especially when using Enterprise, can hurt. A lot. It's really set up for scale-up architectures (and Windows clustering could stand to be improved quite a bit), and if you dare deviate from that notion, you get hurt rather badly.
The problem wasn't finding a replacement for MSSQL, but a replacement for all the tools that come bundled. SSIS, MSAS, SSRS are just a few of the products that make MSSQL a powerful tool. The ability to build and process cubes in MSAS that can be browsed as pivot tables using an Excel plugin made data analysis easily accessible to the analysts at my old company.
I agree that licensing is outrageous and I tried hard to move us onto something else, but replacing the other products would have taken too much time.
Administration, monitoring, maintenance, and such, is all like, yeah its done.
Really I read these comments here about SQL and DB on other platforms and I cannot fathom it, I simply never experience it. The AIX/Oracle gang has half a dozen people or more doing the same work as I for their platform and I don't see the benefit.
However, IBM licensing was significantly more expensive than our SQL Server licensing. As for FTEs, the same DBAs managed DB2 and SQL Server simultaneously whereas we needed an entire new cohort to deal with Oracle.
Although I'm a true-blue PostgreSQL believer, I give major props to the SQL Server team. In comparison, Oracle is a disaster (I was reluctant/forced Oracle 8i/9i DBA, too).
https://kevinclosson.wordpress.com/2012/11/04/oracle-exadata...
As it is, it just seems like two children throwing sand at each other.
Definitely a different tone than what MSFT normally has, but they're confident, and in a world of 7 figure license deals, it's motivation and tone setting for re-selling partners and field staff.
page 1: https://webcache.googleusercontent.com/search?q=cache%3Ahttp...
page 2: https://webcache.googleusercontent.com/search?q=cache%3Ahttp...
I have to say, the tone of both articles seems strangely...petty, for such huge enterprise organisations. Odd.
Personally I like seeing facts not just words. If you're going to be aggressive about your product, fine by me, but.... where's the proof? Where are the un-bias comparison graphs? What use case are you using to prove your point? Without data they all come off looking like two politicians.
All this said, I love SQL Server but never used it at a super large scale though. What I prefer most about it are it's tools which seem to be much more user friendly then Oracles. I'm no DBA though.
That said, pretty much every enterprise database vendor will soak you for every cent you have.
Whenever folk tell me that "relational doesn't scale", I explain that it's been scaled up and out for decades. Don't believe me? Ring any Fortune 500 and ask.
... and then ask how much it's costing them. Be prepared for sticker shock.