Beta

Slashdot: News for Nerds

×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Oracle Promises 100x Faster DB Queries With New In-Memory Option

timothy posted about 10 months ago | from the now-report-back-on-the-double dept.

Oracle 174

Hugh Pickens DOT Com writes "ZDNet reports that Oracle's Larry Elison kicked off Oracle OpenWorld 2013 promising a 100x speed-up querying OTLP database or data warehouse batches by means of a 'dual format' for both row and column in-memory formats for the same data and table. Using Oracle's 'dual-format in-memory database' option, every transaction is recorded in row format simultaneously with writing the same data into a columnar database. 'This is pure in-memory columnar technology,' said Ellison, explaining that means no logging and very little overhead on data changes while the CPU core scans local in-memory columns. Ellison followed up with the introduction of Oracle's new M6-32 'Big Memory Machine,' touted to be the fastest in-memory machine in the world, hosting 32 terabytes of DRAM memory and up to 384 processor cores with 8-threads per core."

cancel ×

174 comments

Oracle gains speed (4, Funny)

rossdee (243626) | about 10 months ago | (#44922365)

Especially upwind, but not 100x

still Emirates Team NZ only need to win one more race..to take back the Americas cup

Re:Oracle gains speed (1)

Anonymous Coward | about 10 months ago | (#44922401)

Especially upwind, but not 100x

still Emirates Team NZ only need to win one more race..to take back the Americas cup

Is it a sign of massive greed when one of the first slashdot comments is about the vendors multi-million dollar boat?

At first glance, one would think "Oracle" is a company devoted to catering high end golf outings and boat racing.

They make software, right? Oh yeah, of course they do, it's this budget line the size of seven IT salaries, how could I forget...

Re:Oracle gains speed (4, Insightful)

Chrisq (894406) | about 10 months ago | (#44922435)

At first glance, one would think "Oracle" is a company devoted to catering high end golf outings and boat racing.

They make software, right?

Only as a means of raising the money for high end golf outings and boat racing.

Re:Oracle gains speed (4, Funny)

ae1294 (1547521) | about 10 months ago | (#44922791)

At first glance, one would think "Oracle" is a company devoted to catering high end golf outings and boat racing.

They make software, right?

Only as a means of raising the money for high end golf outings and boat racing.

With blackjack and hookers....

Re:Oracle gains speed (4, Funny)

ooshna (1654125) | about 10 months ago | (#44922819)

In fact forget the boat racing and the blackjack.

Re:Oracle gains speed (1)

mwvdlee (775178) | about 10 months ago | (#44922913)

You still want to keep the high end gold outings, though?

Re:Oracle gains speed (1)

mwvdlee (775178) | about 10 months ago | (#44922917)

golF. Not that it's a matters much to the cost.

Re:Oracle gains speed (1)

ae1294 (1547521) | about 10 months ago | (#44923313)

In fact forget the boat racing and the blackjack.

I think it's funny the man uses wind power to propel his boat when he runs a fucking TECH company... Seems odd if you ask me. Does he wear a skipper hat or a Militarist uniform?

Re:Oracle gains speed (1)

idontgno (624372) | about 10 months ago | (#44923927)

Seems odd if you ask me. Does he wear a skipper hat or a Militarist uniform?

You will not speak disrepectfully of El Presidente in this way.

We're keeping an eye on you.

-- Oracle Dignity Guard

Re:Oracle gains speed (0)

Anonymous Coward | about 10 months ago | (#44922907)

Oh how much I pine for simpler times, when all we had to do was care about were blackjack and hookers.

BTW, WTF w/ MySQL?

Larry hates the little people so much, because he sure ain't showin' no love. Not even a dog-bone. ...and don't even get me started on OpenOffice.

Re:Oracle gains speed (1)

ae1294 (1547521) | about 10 months ago | (#44923267)

...and don't even get me started on OpenOffice.

Open-office is DEAD to me... Never again... Libra-office has some issues but less than MS WORD if you ask me. It can't do everything WORD can do but who the fuck wants all that stupid shit. If you are a publisher you have to use something better anyway because of the various formating options now... If they aren't then I give them 10 years tops before they die...

Re:Oracle gains speed (1)

ae1294 (1547521) | about 10 months ago | (#44923275)

O... bad form and all but I have to get this out.. FUCK THE CLOUD!

Re:Oracle gains speed (1)

armanox (826486) | about 10 months ago | (#44923305)

In case you didn't notice they don't have OO anymore.

Re:Oracle gains speed (1)

ae1294 (1547521) | about 10 months ago | (#44923339)

Nope I didn't notice and don't care...

Re:Oracle gains speed (0)

Anonymous Coward | about 10 months ago | (#44922809)

You forgot Hawaiian islands.

Re:Oracle gains speed (1)

Anonymous Coward | about 10 months ago | (#44922613)

They make software, right? Oh yeah, of course they do, it's this budget line the size of seven IT salaries, how could I forget...

The size of seven IT salaries? You mean like seven times the salary budget for the entire IT staff, right.

Re:Oracle gains speed (2)

Joce640k (829181) | about 10 months ago | (#44922579)

Anybody know how much this costs?

(It's Monday morning and I need a good laugh...)

Re:Oracle gains speed (0)

Anonymous Coward | about 10 months ago | (#44922647)

Let me guess....

more than you would earn in this and the next lifetime?

Re:Oracle gains speed (2)

ae1294 (1547521) | about 10 months ago | (#44922801)

You only live once but well I hear if you goto hell you program in Fortran all day everyday, forever... They phased out the COBOL because it was much too evil.

Re:Oracle gains speed (3, Interesting)

pinkstuff (758732) | about 10 months ago | (#44922737)

An estimated $100m USD [time.com] . The winner sets the rules for the next one, so if New Zealand wins they will lower the cost, allowing more teams to able to compete.

Re:Oracle gains speed (1)

Rosco P. Coltrane (209368) | about 10 months ago | (#44922633)

Well, they can always cheat [bizjournals.com] to go faster...

Re:Oracle gains speed (1)

Big Hairy Ian (1155547) | about 10 months ago | (#44923145)

Oracle CEO promises faster than light drive

NASA promises to stop using Russians for manned space flights

Which of these is more believable?

Re:Oracle gains speed (1)

armanox (826486) | about 10 months ago | (#44923317)

If Larry Elison thought it would be profitable, we'd get FTL drive....

Re:Oracle gains speed (1)

Anonymous Coward | about 10 months ago | (#44923387)

Is this IBM's Data In Memory (DIM) mantra recycled and reworded?
BLU and SAGE?
Not logging = with ur; Nothing new here, I reckon this idea is 20 years old at least.

This comes on the heels of (0)

Anonymous Coward | about 10 months ago | (#44922385)

This article:

Will Facebook, Twitter, LinkedIn Stay With MySQL? [slashdot.org]

Ellison is doing some public customer relationship management, and I guess this event comes along very timely.

Beowulf (2)

shri (17709) | about 10 months ago | (#44922405)

" hosting 32 terabytes of DRAM memory and up to 384 processor cores with 8-threads per core. "

Let me be the first to point out the Beowulf possibilities with a few hundred of these clustered together. :)

Re:Beowulf (1)

kh31d4r (2591021) | about 10 months ago | (#44922489)

But does it run on Solaris?

Re:Beowulf (5, Funny)

Anonymous Coward | about 10 months ago | (#44922787)

But does it run on Solaris?

Sadly, no. In fact it won't even function as a Minecraft server without some patches and a Java update. I hear that Oracle is still waiting on the vendor for Java update.

Re:Beowulf (2)

sjwt (161428) | about 10 months ago | (#44922495)

Finally a cluster that can Slashdot slashdot.org

Re:Beowulf (2)

ae1294 (1547521) | about 10 months ago | (#44922813)

Finally a cluster that can Slashdot slashdot.org

Not running oracle it can't!
{I never worked with it. They seemed to much like a drug dealer when they came around, they had a building here in SE-VA in Chesapeake}

Re:Beowulf (0)

Anonymous Coward | about 10 months ago | (#44922583)

> Let me be the first to point out the Beowulf possibilities

Not very good ones. Why would you cluster these with commodity hardware available that you can tweak more generously? If you're just trying to make a throwback comment to the days when you didn't have the availability of hardware, that we have now, you might as well talk about how putting together oracle yachts could make an impressive navy. Both comments are equally ignorant of purpose and comedic value. I award you no karma, amateur.

Re: Beowulf (0)

Anonymous Coward | about 10 months ago | (#44922707)

Bloody coward

Re: Beowulf (1)

ae1294 (1547521) | about 10 months ago | (#44922841)

I wanna see a Beowulf of Arduino ATmega328 chips. They cost 4 bucks!

The high-performance Atmel 8-bit AVR RISC-based microcontroller combines 32 KB ISP flash memory with read-while-write capabilities, 1 KB EEPROM, 2 KB SRAM, 23 general purpose I/O lines, 32 general purpose working registers, three flexible timer/counters with compare modes, internal and external interrupts,serial programmable USART, a byte-oriented 2-wire serial interface, SPI serial port, 6-channel 10-bit A/D converter (8-channels in TQFP and QFN/MLF packages), programmable watchdog timer with internal oscillator, and five software selectable power saving modes. The device operates between 1.8-5.5 volts. By executing powerful instructions in a single clock cycle, the device achieves throughputs approaching 1 MIPS per MHz, balancing power consumption and processing speed.

Great news for NSA (4, Funny)

Anonymous Coward | about 10 months ago | (#44922413)

With increasing surveillance on American citizens such database will provide security forces with instant profile of each person. Let's combine that with license plate scanning, cell phone tracking, sexual preferences and health records.

Now we can sleep well at night, our children are safe.

Re:Great news for NSA (1)

NatasRevol (731260) | about 10 months ago | (#44923577)

Not sure why you got modded as funny.

When it's scary as fuck.

Re:Great news for NSA (1)

leuk_he (194174) | about 10 months ago | (#44923859)

If you not living in the amerikas but in the United states of europe, It is so unthought of, it is funny.

But wait... what is that camera doing there....

One Big Memory Machination (5, Funny)

VortexCortex (1117377) | about 10 months ago | (#44922425)

"Big Memory Machine"... So, they finally built Deepthought?

In-memory IO is grand, when that's your're bottleneck. Mine tends to be in the network level, so I use a local daemon for query result caching at the application level as "in-memory" speedup. The speedups are nice, but pricey. Color me unimpressed -- that's pink, BTW; I'm a Caucasoid your colors may vary, but only up to VARCHAR(20);

Uhg. Is "in memory" now just another buz-word? I guess we've come full circle back to Mainframe? Big memory banks are faster and better for a while, but then the bandwidth goes up and the price, reliability and scalability will favor distributed systems (as currently). I wonder which phase of the cycle quantum computing will favor: distributed / localized? You have to take into consideration your user distribution too...

So, eventually you'll want a hybrid system where the memory is distributed and cloned at each query-able interface, but still maintaining the entire dataset "in memory"...
SELECT * FROM earth WHERE answer LIKE "everything";
...
42 rows returned

Re:One Big Memory Machination (0)

Anonymous Coward | about 10 months ago | (#44922465)

You forgot to join on universe and life...

Anyway... Is this not just a new feature to facilitate some extra licensing models...

Re:One Big Memory Machination (1)

knightghost (861069) | about 10 months ago | (#44922531)

Distributed processing requires cache duplication and parallel problems. ERP systems don't follow that. It's the main reason RAC failed.

Re:One Big Memory Machination (3, Funny)

Anonymous Coward | about 10 months ago | (#44922695)

If you used ECC memory, the answer wouldn't have been 42.

Re:One Big Memory Machination (0)

Anonymous Coward | about 10 months ago | (#44922711)

"your're"? that's a new one...

Re:One Big Memory Machination (0)

Anonymous Coward | about 10 months ago | (#44922837)

Color me unimpressed -- that's pink, BTW; I'm a Caucasoid your colors may vary, but only up to VARCHAR(20);

I'm a 'semi-translucent American' so I was just truncated, you inconsiderate clod!

SELECT * FROM earth WHERE answer LIKE "everything";

Was this "LIKE "everything"; " a typo?
- you can speed it up around 100x if you used "= 'everything';"
- you can get more results from "LIKE '%everything%';"

Re:One Big Memory Machination (1)

ae1294 (1547521) | about 10 months ago | (#44922855)

I wonder if all that RAM has ECC or not... I am betting not....

Re:One Big Memory Machination (0)

Anonymous Coward | about 10 months ago | (#44923303)

What happens to this database when you pour ACID on it?

Re:One Big Memory Machination (1)

ae1294 (1547521) | about 10 months ago | (#44923365)

What happens to this database when you pour ACID on it?

Ummm I guess it gets high and starts handing out random information it has stored. Quick someone slip the NSA computers some LSD!

Re:One Big Memory Machination (3, Funny)

doomsayerxero (448334) | about 10 months ago | (#44923747)

That should be Varchar2(20).

Again (2)

hebertrich (472331) | about 10 months ago | (#44922437)

As long as they keep with the extorsion techniques Oracle is famous for they can keep their hardware.

Re:Again (1)

Charliemopps (1157495) | about 10 months ago | (#44922679)

Yea, no shit... is it License renewal time for anyone else right now? Oracles the fucking devil.

Like Microsoft SQL Server (1, Interesting)

Anonymous Coward | about 10 months ago | (#44922455)

Wow, just catching up to Microosft. like SQL Server In-Memory OLTP http://technet.microsoft.com/en-us/library/dn133186(v=sql.120).aspx
And Column Store Indexes released 2 years ago. http://blog.sqlauthority.com/2011/10/29/sql-server-fundamentals-of-columnstore-index/

Re: Like Microsoft SQL Server (-1, Troll)

Anonymous Coward | about 10 months ago | (#44922513)

You are either clueless, high, or both if you think Oracle has to catch up with MS SQL in any way, shape, or form.

Re: Like Microsoft SQL Server (1)

Tridus (79566) | about 10 months ago | (#44922557)

I dunno, Microosft can make a decent .net provider for their database. Oracle has all kinds of catching up to do in that department.

Re: Like Microsoft SQL Server (0)

Anonymous Coward | about 10 months ago | (#44922571)

You're clueless if you don't think in-memory is catching up.

Re: Like Microsoft SQL Server (0)

Anonymous Coward | about 10 months ago | (#44922847)

Ah, one of the "Oracle is better at everything forever and ever and ever and ever" people. I know it's hard to keep up on database trends but if, by chance, you are someone who claims to be a database professional then you might want to do a better job of keeping up.

Re: Like Microsoft SQL Server (0)

Anonymous Coward | about 10 months ago | (#44923455)

Thanks for your input. I am only a DBA for the world's second-largest pharmaceutical company. I deal with dozens of Oracle, Microsoft, and IBM databases daily. The systems that have the highest activity and run our most critical systems are always on Oracle. No exceptions.

Re: Like Microsoft SQL Server (0)

Anonymous Coward | about 10 months ago | (#44923687)

If AC uses Oracle for everything, that for sure means its the only system that can handle it and thus proves it best by default.

Re: Like Microsoft SQL Server (0)

Anonymous Coward | about 10 months ago | (#44923323)

Hahhaha...I get modded down for disrespecting SQL Server? On Slashdot, no less... My god are you people delusional.

Re:Like Microsoft SQL Server (2)

homb (82455) | about 10 months ago | (#44922589)

I don't think it's in the same ballpark. The SQL Server column store seems to be purely for read-only:

Keep in mind that once you add a column store to a table, though, you cannot delete, insert or update the data – it is READ ONLY.

That's nowhere near the complexity of what Oracle is doing, simultaneously providing both a row and column based access to the data. Not that I think this is a good thing, I don't. In most cases you're much better off using a kickass columnar db and handling the batch updates from the upstream app servers. When you plan for building a col-based architecture, you can be much more efficient. Just look at kdb & co.

Re:Like Microsoft SQL Server (3, Interesting)

SDrag0n (532175) | about 10 months ago | (#44922825)

Re:Like Microsoft SQL Server (0)

NatasRevol (731260) | about 10 months ago | (#44923595)

LOL at Microsoft saying they'll get to something 'next year'.

This merely allows poor code to suck less. (4, Interesting)

pla (258480) | about 10 months ago | (#44922461)

First, let me say that I would love to have a table option to keep a particularly heavily-hit table always in memory.

This ain't it.

From TFA, "Maintaining those indexes is expensive and slows down transaction processing. Let's get rid of them," Ellison remarked. "Let's throw all of those analytic indexes away and replace the indexes with in-memory column sort."

This merely minimizes the penalties of poor indexing and RBAR by making complete table scans on arbitrary columns faster. Apparently Mr. Ellison has forgotten his algoithmics and combinatorics - Oh, wait, no he didn't, he dropped out as a sophmore. Pity, because had he stayed, he would have learned that even with a 1000x slower storage medium, an O(log N) algorithm (index seek) will eventually beat an O(N log N) algorithm (column sort).

Thanks, Larry, but you want to make Oracle faster? Remove cursors from the core language, and although that alone won't "fix" it, you'll see all the hacks who can't think in set-based logic drop out overnight.

Re:This merely allows poor code to suck less. (3, Informative)

homb (82455) | about 10 months ago | (#44922621)

From TFA, "Maintaining those indexes is expensive and slows down transaction processing. Let's get rid of them," Ellison remarked. "Let's throw all of those analytic indexes away and replace the indexes with in-memory column sort."

This merely minimizes the penalties of poor indexing and RBAR by making complete table scans on arbitrary columns faster. Apparently Mr. Ellison has forgotten his algoithmics and combinatorics - Oh, wait, no he didn't, he dropped out as a sophmore. Pity, because had he stayed, he would have learned that even with a 1000x slower storage medium, an O(log N) algorithm (index seek) will eventually beat an O(N log N) algorithm (column sort).

I think you misunderstand the way columnar databases work. They are not doing a column sort the way you think. The column itself is an index.
Of course the inanities coming out of Ellison's mouth don't help explain things correctly. No Larry, you don't do away with indexes. You mostly store indexes on everything, automatically.

Thanks, Larry, but you want to make Oracle faster? Remove cursors from the core language, and although that alone won't "fix" it, you'll see all the hacks who can't think in set-based logic drop out overnight.

Can't argue there!

Re:This merely allows poor code to suck less. (3, Insightful)

oranGoo (961287) | about 10 months ago | (#44922789)

From TFA, "Maintaining those indexes is expensive and slows down transaction processing. Let's get rid of them," Ellison remarked. "Let's throw all of those analytic indexes away and replace the indexes with in-memory column sort." This merely minimizes the penalties of poor indexing and RBAR by making complete table scans on arbitrary columns faster. Apparently Mr. Ellison has forgotten his algoithmics and combinatorics - Oh, wait, no he didn't, he dropped out as a sophmore. Pity, because had he stayed, he would have learned that even with a 1000x slower storage medium, an O(log N) algorithm (index seek) will eventually beat an O(N log N) algorithm (column sort).

RTA - the improvement is there specifically for real time analytic workloads. In these kind of workflows the optimal algorithm is O(n) in general case and indexes are useless (query optimizing engine will always choose scans as you need to visit a lot of data). You might know a thing or two about algorithms, but you should brush up on problem analysis 101.

Other mistakes in logic: Index seek and column sort are not different algorithms for the same task so comparing them brings little insight (without considering some other details of the query optimizer). This leads you to nonsensical claim that O(log N) will eventually beat or be equal to O(N log N). It is not eventually, the first will be always faster or equal.

Re:This merely allows poor code to suck less. (2, Informative)

Anonymous Coward | about 10 months ago | (#44922979)

This leads you to nonsensical claim that O(log N) will eventually beat or be equal to O(N log N). It is not eventually, the first will be always faster or equal.

Er... no, not if you think about constant terms. Consider 8 * (log N) + 3 > 2 * ( N log N) for N=4 for example. Big O notation is about the asymptotic performance for large N. It's quite common that the most efficient algorithms for large data sets may provide no benefit for trivial cases and may in fact be slower.

Re:This merely allows poor code to suck less. (1)

Jorl17 (1716772) | about 10 months ago | (#44923481)

woosh!

HOW MUCH YOU WANNA BET !! (0)

Anonymous Coward | about 10 months ago | (#44922483)

Some of that memory corrupts, Shhhh !! Who is going to know !! Back in the day INT2 would save our asses but that is no more !! Today it is better to let the errors creep than to report !! And no, ECC does not save itself !! ECC lets vendors sell defective RAM !! Bet your life on that !!??

32 TB is 256 T bits which is 2 ^

Re:HOW MUCH YOU WANNA BET !! (0)

Anonymous Coward | about 10 months ago | (#44922931)

32 TB is 256 T bits which is 2 ^

What a turn off. I'll wait for the 64 TB version, TYVM. Hell, even the new iPhone is 64 bit now ;-)

Word Processor and Reader for Microsoft Office. (-1, Offtopic)

mustafawi (3177003) | about 10 months ago | (#44922505)

Documents - Word Processor and Reader for Microsoft Office.http://bit.ly/DWPRM By Irfan Farooqi Lightweight office work on the go Backup of documents Quick access to Documents, Spread sheets, Presentations, notes and memos word processing Pocket Spreadsheet Pocket Presentation Download : https://itunes.apple.com/us/app/documents-word-processor-reader/id642314248?mt=8 [apple.com]

Dual format (0)

Anonymous Coward | about 10 months ago | (#44922529)

Finally, the reason Oracle requires joining to dual is revealed.

OTLP? (1)

tonywestonuk (261622) | about 10 months ago | (#44922543)

whats OTLP?

Re:OTLP? (0)

Anonymous Coward | about 10 months ago | (#44922591)

Let's see:
TLA = Three Letter Algorithm
OTP = One Time Pad
OLPC = One Laptop Per Child
therefore
OTLP = One Three Letter PerChildLapTopPadAlgorithmWithOffByOneError

Re:OTLP? (1)

wmac1 (2478314) | about 10 months ago | (#44922703)

How about using google to find out?

http://datawarehouse4u.info/OLTP-vs-OLAP.html [datawarehouse4u.info]

Sorry, but isn't this just SAP HANA? (1)

yooy (1146753) | about 10 months ago | (#44922569)

Re:Sorry, but isn't this just SAP HANA? (1)

Anonymous Coward | about 10 months ago | (#44922845)

No, SAP HANA is just an in-memory database.

This sounds more like Teradata [wikipedia.org] , but in-memory.

Or Sybase IQ [wikipedia.org] , if you want to stay in the SAP stable.

Oracle-friendly site(s) (2)

SigmundFloyd (994648) | about 10 months ago | (#44922587)

14 hours ago, itnews.com.au runs a story (promptly picked up by /.) about how the social networks are staying with MySQL [itnews.com.au] . In the article, it is suggested that the switch to MariaDB by some Linux distros is a "political move", and that Google's switch might be a retaliation against an unrelated lawsuit from Oracle. Also, it's mentioned (twice, with the same wording) that the Mozilla foundation is "upgrading from MariaDB to MySQL 5.6" (emphasis added).

7 hours ago, itnews.com.au runs a story (promptly picked up by /.) about how Oracle's 12C database will be 100x faster, despite the fact that we only have Oracle's CEO word for it.

Now that's what I call an Oracle-friendly site (or two?)

Re:Oracle-friendly site(s) (1)

bluefoxlucid (723572) | about 10 months ago | (#44922831)

MySQL is a piece of shit. MongoDB and PostgreSQL, depending on the data layout (MongoDB if you're looking for an indexed XML-YAML-JSON type of data with reliable data integrity in a cluster; PostgreSQL if you're looking for an indexed CSV with good replication and only minimal probability of basically silent data loss in a fail-over scenario).

Re:Oracle-friendly site(s) (1)

citizenr (871508) | about 10 months ago | (#44923261)

MongoDB
reliable
data integrity
in a cluster

You mean if you run 3 copies of same database and constantly compare them with each other so in case of discrepancy you can at least guess which one is correct?

32TB of RAM = Everest-sized UPS (2)

Rosco P. Coltrane (209368) | about 10 months ago | (#44922603)

Just you hope Oracle maintains the batteries properly, especially since an emergency save-to-disk is going to take more than a few minutes...

Re:32TB of RAM = Everest-sized UPS (1)

Charliemopps (1157495) | about 10 months ago | (#44922709)

lol, you should see our battery room. We've got a 10 story building and every single computer in it is on the UPS. And that is just there to keep everything up long enough for the two 500hp natural gas generators can kick in. They feed off both the city feed as well as have their own separate tanks. Short of a meteor hitting the building the power will never go out. And we're pretty much small time.

Re:32TB of RAM = Everest-sized UPS (2)

KiloByte (825081) | about 10 months ago | (#44922739)

Short of a meteor hitting the building the power will never go out

A small fire in a strategic place... Or perhaps, just some minor failure of something you thought was separated enough.

Re:32TB of RAM = Everest-sized UPS (1)

mjwalshe (1680392) | about 10 months ago | (#44922853)

Maybe not do they have mo e than one feed for the grid? though this sounds like a telco exchange set up (central office for you colonials)

Re:32TB of RAM = Everest-sized UPS (1)

Anonymous Coward | about 10 months ago | (#44923561)

my favorite is when the UPS gets input power in a waveform it just doesn't quite like, and instead of going to straight battery--or conditioning the power, it just turns everything off.

Re:32TB of RAM = Everest-sized UPS (1)

dbIII (701233) | about 10 months ago | (#44922783)

two 500hp natural gas generators can kick in.

Two is definitely a good idea. One workplace I was at had a 20MW generator that was started for testing every month for thirty years without fail and then just could not get going the day it was needed.

Re:32TB of RAM = Everest-sized UPS (0)

mjwalshe (1680392) | about 10 months ago | (#44922833)

you are testing your gensets every day right

Re:32TB of RAM = Everest-sized UPS (1)

ae1294 (1547521) | about 10 months ago | (#44922887)

I'm so glad you just said that...
A pox upon your data center! Curses, unknowable horrid darkness will be at your throat, unseen and without form.... Soon... May it consume your data and your soul for my Master waits for the...

Re:32TB of RAM = Everest-sized UPS (1)

fuzzyfuzzyfungus (1223518) | about 10 months ago | (#44922763)

Just you hope Oracle maintains the batteries properly, especially since an emergency save-to-disk is going to take more than a few minutes...

Obviously, if you are in the market for a machine with 32TB of RAM, your problem is probably important and/or expensive enough to not skimp on (relatively) cheap details like backup power; but, so long as you were willing to stop what you were doing and dump to disk, 32TB wouldn't be too bad.

Magnetics can't do random I/O for shit; but even the cheap consumer crap plays surprisingly nicely with linear reads or writes (high areal density will do that for you).

You would never actually want to shut down such a beast (even aside from whatever the downtime costs you, pure depreciation probably costs more per hour than the team running the thing); but doing a straight linear dump to HDD would be totally doable on some fairly prosaic storage hardware.

Re:32TB of RAM = Everest-sized UPS (1)

dbIII (701233) | about 10 months ago | (#44923183)

Here's one supplier of many that can feed a mountain of such machines
http://cfaspower.com/11MW-20MW_CTG.html [cfaspower.com]
Note the age of some of the things. This stuff is not rocket science, it's 1950s jet engine science!

Obligatory (1)

SuricouRaven (1897204) | about 10 months ago | (#44922611)

Now if we just put a few of those in a beowolf cluster...

Of course it's faster (0)

Anonymous Coward | about 10 months ago | (#44922623)

Oracle is closed source and developed by professionals. Take a hint, MySQL.

Re:Of course it's faster (0)

Anonymous Coward | about 10 months ago | (#44922889)

er, MySQL is part of Oracle these days

384 CPUs equal to ...? (0)

Anonymous Coward | about 10 months ago | (#44922649)

Is it 4 or 8 i7s?

Re:384 CPUs equal to ...? (1)

robthebloke (1308483) | about 10 months ago | (#44923043)

= the GPU on an AMD A10

MonetDB/X100 and Vectorwise (2)

Skinkie (815924) | about 10 months ago | (#44922725)

Columnstore databases such as MonetDB and their commercial spinoff Vectorwise [wikipedia.org] (now Actian) already showed [tpc.org] this can be achieved with open source and proprietary code.

Failzor5! (-1)

Anonymous Coward | about 10 months ago | (#44922765)

Over a ,quaLity

Did something a LOT like this in 1996 (0)

Anonymous Coward | about 10 months ago | (#44922779)

For EEC Systems/SuperSpeed.com & it placed as a finalist 2 yrs. in a row @ MS TechEd 2000-2002 in its hardest category: SQLServer Performance Enhancement (by placing the db devices & tables (temp too)) into RAM via the SuperDisk II program (which I also increased SuperCache II's effectiveness by up to 40% for also) & it also reviewed VERY POSITIVELY in Windows IT Pro magazine that same 1st year 1996 (then Windows NT Magazine) - & yes, for performance of course, it works!

* Pretty much "goes without saying" that IF/WHEN you avoid diskbound latencies, you're going to speed up!

(Is THIS precisely the same? Not quite - they're distributing the data across diff. systems from what I read/iirc, & eliminating indexing - which is the ONLY part I question here...)

APK

P.S.=> "What has been will be again, what has been done will be done again; there is nothing new under the sun." - Ecclesiastes 1:9 ...

... apk

Re:Did something a LOT like this in 1996 (0)

Anonymous Coward | about 10 months ago | (#44922985)

Hey, APK, remember to save this as a reference to "prior art" when Oracle wants to patent their idea.

"There's very little original thought..." (0)

Anonymous Coward | about 10 months ago | (#44923331)

I state that since Mr. John Enck & I BOTH came up with the idea independently, specifically for DB work, & posted it to their "FAQ" section of the EEC Systems/SuperSpeed.com website in fact - I was doing that type of thing with DBase III temp/scratch tables way, Way, WAY before it also albeit native ramdisk softwares (DOS usually) -> http://ep.yimg.com/ty/cdn/superspeed/ScReadMe41.txt [yimg.com]

(Hence the biblical chapter/verse quote I closed off the post you just replied to, since imo, it too "holds true" here)

I suspect not ONLY for Oracle doing pretty much what I did, but very possibly OTHERS DOING IT BEFORE MYSELF doing it too!

HOW/WHY?

Heck: Since I am FAIRLY certain I was "no 1st" here because it's SO DAMNED OBVIOUS to do for performance purposes per less latencies in RAM vs. being diskbound... (but, you never know!).

---

* I do, however, also see a LOT of usage of in-memory DB tech, Virtual Machine, Terminal Server applications the past 1/2 decade now in "industry" via SSD usage, which amounts to damn near the SAME thing too - & it seems the "youth of today" have stumbled upon it, finally, lol!

(Which is yet later another way I applied this, not with software ramdisks in software this time, but later, in hardware via CENATEK's "RocketDrive" PC-133 SDRAM 4gb/16gb PCI 2.2 bus striped-spanned hardware ramdrive cards, & later still doing the same with Gigabyte IRAM 4gb/16gb striped-spanned DDR-2 RAM SATA II bus PCI Express based ramdrive boards - this idea/technique's "taken off" most recently though on the areas noted, using FLASH based SSD's... the only REAL difference, but the principals are the same - albeit FASTER in system memory though using ramdrive softwares!).

---

Thus - I really don't *THINK* Oracle CAN patent this technique to be quite honest... why? Well, what the "artsy/fartsy" people are wont to state, in "It's been done..."

---

Heh - This ALL reminds me of the "Outer Limits" episode "The FINAL exam" - when Seth the main character says:

"When a Science is ready? It can't help but make the next discovery ..."

APK

P.S.=> Bottom-Line: Yes - I'd be VERY surprised if my idea came before anyone else on this note on the entire planet & mainly, since BOTH myself + Mr. Enck (technical editor for Windows IT Pro/Windows NT mag) both put it up on their website as a technique for superior performance purposes (albeit I had been applying it & told the folks @ EEC/SuperSpeed to do so too, on paid contract, while increasing SuperCache I/II's efficacy by up to 40% in code too)... apk

Re:Did something a LOT like this in 1996 (1)

Virtucon (127420) | about 10 months ago | (#44923497)

IBM had something like this in Mainframes back in the 80s with DB2 and Hyperspaces. Back then I recall management wondering what they'd do with it.

I'll have one someday. (1)

eclectro (227083) | about 10 months ago | (#44923457)

Ellison followed up with the introduction of Oracle's new M6-32 'Big Memory Machine,' touted to be the fastest in-memory machine in the world, hosting 32 terabytes of DRAM memory and up to 384 processor cores with 8-threads per core."

This should be on my desk in about 4 years.

Okay, fast database - high transaction rate, BUT (0)

new death barbie (240326) | about 10 months ago | (#44923791)

...how do you replicate this to an offsite hot standby, so that you don't look stupid when the power goes out? Or whatever other disaster occurs, because you just KNOW a single basket full of this many eggs is going to attract tornadoes, Cat 5 hurricanes, earthquakes, and meteors. Also coffee spills, hungry rodents, and burst pipes.

Also I believe OTLP == Oracle Typical Ludicrous Pricing

Load More Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Create a Slashdot Account

Loading...