Re:Good thing databases are perfect!
on
Databases and Privacy
·
· Score: 2, Interesting
And, it seems that only reasons are to prevent a few people from blowing things up, and to sell us more razorblades.
... and so "they" can all trash our Resume because some data entry guy at AT&T screwed up and sends you a phone bill for $5000 which you don't pay so they took you to Court and Felonise you, great system!
Is the recession really caused by a crash in IT, or because 99% of people in the US are now suddenly unemployable because our great grandmothers had cancer or something? Health insurance kills people that are poor and/or need expensive treatment, and now people with a 1% possibility of health problems in future will be denied jobs. Seems like Hitler's eugenics program is well underway in the US.
the day is rapidly approaching when the internet will be like the current telephone or snail mail networks. Actually, it'll probably subsume them, and adopt most of their regulations. Then you'll have a collection of national intranets, with the internet being the scheme which negotiates communication among them
You are correct, I value the Internet far more than McDonalds and yet I spend $8 a week on Big Macs because they provide a convenient service which I actually pay for and consume.
Normal customers demand access to the Internet for 3 reasons:
To demand content from servers on the Internet (worldwide commons)
To publish content (to the worldwide commons) acting as a server
Combination of 1 and 2
McDonalds make money from me because they control their product until I pay them (I get Big Mac AFTER I PAY them). My ISP makes money from me because I am willing to PAY MONEY for two-way access to the worldwide commons. This is where it breaks down - people worlwide are not willing to pay servers for access to their content, and the speed we're used to getting stuff over the Internet doesn't allow us to whip out our credit cards and pay for access to any website/Kazaa server. What's needed is a mechanism whereby ISPs will increase their fees and hold a trust fund which will be donated to accessed ISPs and/or webservers logged by the ISP's packet counters (if they can run carnivore, then they should be able to do this).
Your ISP pays you a small amount of money for your outgoing content requaested by incoming non-DDoS packets. Next webservers will block all IP addresses coming from ISPs that don't pay them for their content or generate fake hits. These blocked IP addresses will not get content and will change ISPs, that way Capitalism can be extended to the Internet.
Think about it - the only difference between Slashdot and a blank IP address is a huge amount of Slashdot content requested from Slashdot's Port 80 to ISPs. Slashdot should be allowed to bill these ISPs for content requested. The IETF must alter all Internet protocols accordingly.
He could even spy on his employees from the deck of the ship: He brought up Global Positioning System maps that showed him the precise location of each of his trucks, down to the intersection. If an employee was off-track, he could fire off a text message to the truck
Our computing technology is being usurped by Managers to imprison employees. Linux makes slavery cheaper than Micro$oft does. Is this something to be proud of?
Globalisation will have the general effect of allowing cheaper imports of all sorts of products. The US economy produces teh most high tech stuff in the world which is why it tries to patent everythign in sight
As for being stubborn, maybe a little. I'm interested in people proving what they say. The arguments about cellphones can most certainly be proven quantitively and should be. Instead of asanine remarks along the lines of "because my monitor shakes, the phone MUST produce 200W." It's foolish to draw such a conclusion. I'd have to think that you would agree being of a learned background
You are incorrect. To be truly true to a learned background, and to be honest instead of banging you over the head with such a background, the truth must be stated, and that is that it's in no significant body's interests to perform research.
If the Government spent $10 billion on cellphone health research they'd put it all over CNN, just like their extra funds for bioterrorism defences. I'd guess that if Abraham Lincoln was still in power he would perform the research as fast as possible and give us a lecture about irresponsible and premature adoption of new technology
It's not in the drug companies' interests to research whether cellphones are dangerous, just like it's not in their interests to tell people to control calories to lose weight; instead they sell drugs and Slim-Fast to make money off our fatness. If we become less fat, we become healthier and they lose money.
So who would make money from proving that cellphones are dangerous? Uhhhhh, hmmmmmmmm, I can't think of any, now this is the real source of the problem, the citizens are too busy worrying about their dwindling jobs
Calculating the maximum transient output power of the battery is meaningless, maybe the brain works better after being stimulated by weak magnetic fields, God knows. Nobody knows, nobody wants to know, and it's in no megacorporation's interests to know so a PhD electrical engineer can sit and guess just like a structural engineer guessed 30 years ago that nobody will take an aeroplane and deliberately smash it into a skyscraper.
3.6 Volts * 0.9Ah = 3.2 Watt Hours = 3.2 x 60 x 60 Joules = 1152 Joules of energy contained in your battery, Sir
200 Watt 1ms pulse = 200 x 10^-3 Joules = 0.2 Joules of energy in each microwave pulse
So your battery has enough power to generate 5000 200Watt pulses, each 1millisecond in length. My microwave takes ten 5-second 300Watt pulses to defrost my dinner. I suggest you don't argue against someone with a Masters degree in Electronics and Computing in future (sorry to pull intellectual rank but a stubborn brick wall needs a hard head to break it)
But in terms of what Monty said, MySQL should be treated like an Array with some nice features like recovery. It's obvious to any DBA that MySQL wasn't designed for multiple tables with related data. It would be easier to handle linked lists. I'm not saying MySQL is bad - you might as well make Array illegal in which case 99% of all C++ programs would collapse. Use whatever data storage is appropriate.
There's no point MySQL getting transactions, because then it would become JARDB (Just Another Relational Database). Even Micro$oft doesn't copy Oracle's DB precisely despite the fact that Oracle outclasses SQL Server.
MySQL excels at small often-read datasets with unrelated tables. Relate them at your peril! Or just get Postgres. Why do we want to see MySQL just become another Oracle or Postgres clone? Why do we try to turn an ant into a homo sapien?
What you describe is just a prejudice... the "old dogs can't learn new tricks" mentality which is, unfortunately, prevalent in our society
Actually it's quite accurate. I tried to teach my friend's very old dog a new trick but he didn't learn it. Actually he didn't move at all, and he wasn't breathing either.
The advantages of going to grad school, particularly when slightly older, during a recession are numerous. I did it during the last two recessions (MSc in the early eighties, a Ph.D. and a couple postdocs during the early nineties), so I speak from experience
Hmmmmm, so a recession DEFRAGS the economy - it's time to do some intellectual maintenance.
So by the same logic, we low-level formatted the Iraqi Government, hmmmmmmmm
I don't think Monty understands any of this; in the article he seems to say that ACID, rather than being a fundamental principle of relational databases, is just something you need to do because disks are slower than RAM
If you think about it, any Array has all the ACID properties (if you ignore writing out to disk). You'll at least get a runtime error upon ACID failure.
Why is everybody looking for the Killer App just like the Gold Rush? IPv6 is supposed to be a Killer App and yet it's not an App at all, and nobody wants to implement it because they have existing IPv4 and NAT with internal 10Gig Eth backbone.
Come on, seriously, is there going to be a Killer App that is going to make Silicon Valley explode and get convicted murderers with zero experience jobs as C++ software engineers?
INCORRECT! Huge power surges can be created by using inductors, kiloVolts can come from a 1.5V battery to power a Xenon strobe.
Your 600W microwave simulates cooking at 300W when you put it on Defrost NOT by halving the GaAs oscillator's Vds but by running the GaAs at full power for 10 seconds, then stopping it for 10 seconds. In other words, once your food starts cooking, if you remove the heat it doesn't become raw again - it's a cumulative chemical change.
The issue is does the 200W GSM pulse trigger even a small amount of your brain to cook? If so the damage is cumulative.
So you can cook a man's brain with 200W in one second, or you can cook a man's brain with 10,000 x 200W 0.1ms pulses GSM has chosen the latter (unless being a living organism the brain's self-repair can detect and replace the cooked cells, but usually with repetitive damage cells become irreparably precancerous like in Barrett's Sydrome).
on the other hand, an 8GB database would fit very nicely in RAM on a 64Bit computer
Are you out of your mind? All the professionals deal with huge databases on Oracle, and Solid State Disk offers fast storage. People dealing with 8GB databases will be Windowze users and will look at their automatic gear-shift in their car puppy-eyed when you tell them that they lost data in their RAM database because they didn't BACK IT UP!
The very idea of click-click-doubleclick MCSEs giving every Mom and Pop shop a RAM database is scary as hell, lots of businesses will go bankrupt when slack-jawed yokel says, "Why y'all I can here ye speed up all ye databases y'all YEE HA, lookey here it's a called a RAYME-data-er-base YEEEEHA!"
Additionally, after backup/restore/checkpoint/DEFRAG INDEX and memory-intensive queries or stored procedures on the SQL server, the RAM-cache becomes flushed to make room for the operation (LRU pages purged). This is why after some queries/operations the database becomes inexplicably lethargic. Many databases don't even allow the DBA to force the RDBMS to disallow LRU Primary Key main table Indexes to be flushed out of cache! We cannot control cache! This is why many databases warn against running memory-intensive queries (in violation of SQL standard), it's because they use simple caching techniques.
No you won't. There's a difference between a transaction not happening, and a transaction being lost
I am aware of the difference of losing a COMMITed transaction. The RDBMS will return COMMITed when fsync() returns from the transaction log write and/or distributed propogation. I assume RAID-5 array dual HD failure because upon single HD failure it takes 5 minutes to get to the server room, 10 minutes to unpackage a replacement HD, 10 minutes for making sure NOT to pull out the wrong hard disk, 2 minutes for hot-swapping, and 1 hour for RAID-5 data reconstruction. During RAID-5 data reconstruction the strain on the remaining HDs is so massive with constant seeks that their temperature increases to 85 Celsius during data regeneration from RAID-ECC. My failure calculations assume immediate HD failure of all disks above 75 Celsius and therefore all our data will always be destroyed as no RAID array in any configuration can survive simultaneous failure of all hard disks.
RAID controllers must be redesigned to minimise the load on the surviving RAID HDs during reconstruction.
The difficulty with RAM is that it loses its data after you turn off the power.
Why does everyone think this? If your Motherboard cracks in mid-transaction you will lose transactions. Five Lithium batteries (simple mechanics, every watch uses them, rarely fail) connected through diodes in parallel and going through a voltage regulator chip for RAM provides a much more reliable persistant storage than hard disk.
UPS is old technology, the battery needs constant replacement, and very few have multiple redundant batteries and/or transistors to deal with wear and tear. Yes even a simple MOSFET transistor is not 100% reliable. Usually the only way to tell a battery is dead is your UPS fails when you need it (this happened to us when my MD was demonstrating our service live to customers, afterwards was the only time he's taken less than 9 months to sign off a purchase order on new equipment). A UPS also has a power cord to pull out when you recoil after burning your fingers on a Seagate Cheetah 15000RPM HD in the server room. A UPS also trips if you overload it, which again means the UPS fails when you most need it.
Other posts mention cosmic radiation at high altitude makes RAM fail. Last time I checked there were no Quad-Xeon Oracle databases on Concorde, although if the International Space Station were to use one this might pose a problem for non-ECC RAM. Anyway, somebody could always write a driver to do software-ECC with Reed-Solomon for RAM if it becomes necessary.
Huge databases (>500 Gigabyes) would benefit most from this as running a simple OUTER JOIN query on the biggest tables will require most of the database to be called into RAM.
Small databases become slow due to HD latency problems if they do a lot of WRITE operations (the database is stored in RAM, the transaction log is appended to, COMMIT TRANS). This would benefit least FROM RAMdisk because a HD append operation is cheap, however it would benefit database speed in mid-backup
Mid-size databases become HD-intensive due to aggregate queries/triggered operations over large '>RAM' datasets. For instance enforced cascading deletes where millions of tuples are being deleted cascaded to hundreds of other unindexed tables (in my job I go to the toilet whenever I run a query like this).
Huge databases where 'Index size' > 'RAM size' - the simplest query would benefit hugely from more RAM or faster storage or RAM-storage. With current databases this would be a 10Gig Eth connection to a Terabyte RAMSan solid-state disk.
In the future, who knows, maybe a FPGA/ASIC DPU (Database Processing Unit) for INSTANT COMMIT like NVidia's GPU?
Homeworld. Especially the beginning of the 'gardens of Kadesh' level
Yes, truly beautiful and eerie music (don't play this game in the dark!), and the GUI with virtual reality mouse-control amazes my friends every time. "Click the right mouse button, yeah now move the mouse around" and they always go "Whoaaaaaaa!" like in drivers' ed the first time you realise a steering wheel directly controls the car's direction.
Also your racist comments show how little you know about Indian programmers some of who are big names in the American Computer Industry
What's racist?
1. Go the Bank and change $1 for 5000000 Indian Rupees Slightly exaggerated
2. Hire 1000 Indian programmers with above currency Senior managers at HP sign off $1million invoices all the time, any expenditure less than $1million is not questioned.
3. Tell the programmers to recompile all statically-linked applications with the new libraries Easy
4. Hire unemployed American programmer [oddtodd.com] for $20000 to translate the program from Hindi to English Fair enough, with Indian programmers you don't need this step, I retract this particular statement.
5. Charge large corporations big $$$ for upgrading all their software Or less $$$ than competing US corporations with US programmers paid $30000 for just one programmer.
5. Profit!!! (Really)
The CEO of Walmart is richer than Bill Gates, despite the fact the majority of his US workforce is living on charity and Government aid. Your taxpayer dollars are paying for Walmart's reduced salary, and therefore Walmart's profits. Walmart keeps its salaries low by requiring drug tests and by having patrolling managers who are under orders to stop employees talking (time theft) because "idle talk forms unions"
For one, if you statically link your application then anytime there's a security fix or change to the linked library you'll need to recompile the application,
Easily solved, all you have to do is,
1. Go the Bank and change $1 for 5000000 Indian Rupees 2. Hire 1000 Indian programmers with above currency 3. Tell the programmers to recompile all statically-inked applications with the new libraries 4. Hire unemployed American programmer for $20000 to translate the program from Hindi to English 5. Charge large corporations big $$$ for upgrading all their software 5. Profit!!! (Really)
And the faster CPUs gave rise to the OOP paradigm. While it primarily is a nice theoretical concept for safer and more secure program, it's used these days just for code-bloat and GUI overload. Inpedendent studies show that in fact 73 percent of all "OOP" code is just imperative with C++ class bloat added
GUI bloat can be eradicated by removing the User Interface. If your spreadsheet runs as a service with no User Interface then naturally 99% of the code becomes redundant and can be deleted.
I never thought the Osprey (latest military fly-by-wire plane) would need a reset button, and that this reset button would kill people, but it does!
Slashdot Mirror
Is the recession really caused by a crash in IT, or because 99% of people in the US are now suddenly unemployable because our great grandmothers had cancer or something? Health insurance kills people that are poor and/or need expensive treatment, and now people with a 1% possibility of health problems in future will be denied jobs. Seems like Hitler's eugenics program is well underway in the US.
Normal customers demand access to the Internet for 3 reasons:
- To demand content from servers on the Internet (worldwide commons)
- To publish content (to the worldwide commons) acting as a server
- Combination of 1 and 2
McDonalds make money from me because they control their product until I pay them (I get Big Mac AFTER I PAY them). My ISP makes money from me because I am willing to PAY MONEY for two-way access to the worldwide commons. This is where it breaks down - people worlwide are not willing to pay servers for access to their content, and the speed we're used to getting stuff over the Internet doesn't allow us to whip out our credit cards and pay for access to any website/Kazaa server. What's needed is a mechanism whereby ISPs will increase their fees and hold a trust fund which will be donated to accessed ISPs and/or webservers logged by the ISP's packet counters (if they can run carnivore, then they should be able to do this).Your ISP pays you a small amount of money for your outgoing content requaested by incoming non-DDoS packets. Next webservers will block all IP addresses coming from ISPs that don't pay them for their content or generate fake hits. These blocked IP addresses will not get content and will change ISPs, that way Capitalism can be extended to the Internet.
Think about it - the only difference between Slashdot and a blank IP address is a huge amount of Slashdot content requested from Slashdot's Port 80 to ISPs. Slashdot should be allowed to bill these ISPs for content requested. The IETF must alter all Internet protocols accordingly.
Globalisation will have the general effect of allowing cheaper imports of all sorts of products. The US economy produces teh most high tech stuff in the world which is why it tries to patent everythign in sight
You are correct
If the Government spent $10 billion on cellphone health research they'd put it all over CNN, just like their extra funds for bioterrorism defences. I'd guess that if Abraham Lincoln was still in power he would perform the research as fast as possible and give us a lecture about irresponsible and premature adoption of new technology
It's not in the drug companies' interests to research whether cellphones are dangerous, just like it's not in their interests to tell people to control calories to lose weight; instead they sell drugs and Slim-Fast to make money off our fatness. If we become less fat, we become healthier and they lose money.
So who would make money from proving that cellphones are dangerous? Uhhhhh, hmmmmmmmm, I can't think of any, now this is the real source of the problem, the citizens are too busy worrying about their dwindling jobs
Calculating the maximum transient output power of the battery is meaningless, maybe the brain works better after being stimulated by weak magnetic fields, God knows. Nobody knows, nobody wants to know, and it's in no megacorporation's interests to know so a PhD electrical engineer can sit and guess just like a structural engineer guessed 30 years ago that nobody will take an aeroplane and deliberately smash it into a skyscraper.
= 3.2 x 60 x 60 Joules
= 1152 Joules of energy contained in your battery, Sir
200 Watt 1ms pulse
= 200 x 10^-3 Joules
= 0.2 Joules of energy in each microwave pulse
So your battery has enough power to generate 5000 200Watt pulses, each 1millisecond in length. My microwave takes ten 5-second 300Watt pulses to defrost my dinner. I suggest you don't argue against someone with a Masters degree in Electronics and Computing in future (sorry to pull intellectual rank but a stubborn brick wall needs a hard head to break it)
There's no point MySQL getting transactions, because then it would become JARDB (Just Another Relational Database). Even Micro$oft doesn't copy Oracle's DB precisely despite the fact that Oracle outclasses SQL Server.
MySQL excels at small often-read datasets with unrelated tables. Relate them at your peril! Or just get Postgres. Why do we want to see MySQL just become another Oracle or Postgres clone? Why do we try to turn an ant into a homo sapien?
So by the same logic, we low-level formatted the Iraqi Government, hmmmmmmmm
Come on, seriously, is there going to be a Killer App that is going to make Silicon Valley explode and get convicted murderers with zero experience jobs as C++ software engineers?
Your 600W microwave simulates cooking at 300W when you put it on Defrost NOT by halving the GaAs oscillator's Vds but by running the GaAs at full power for 10 seconds, then stopping it for 10 seconds. In other words, once your food starts cooking, if you remove the heat it doesn't become raw again - it's a cumulative chemical change.
The issue is does the 200W GSM pulse trigger even a small amount of your brain to cook? If so the damage is cumulative.
So you can cook a man's brain with 200W in one second, or you can cook a man's brain with 10,000 x 200W 0.1ms pulses
GSM has chosen the latter (unless being a living organism the brain's self-repair can detect and replace the cooked cells, but usually with repetitive damage cells become irreparably precancerous like in Barrett's Sydrome).
The very idea of click-click-doubleclick MCSEs giving every Mom and Pop shop a RAM database is scary as hell, lots of businesses will go bankrupt when slack-jawed yokel says, "Why y'all I can here ye speed up all ye databases y'all YEE HA, lookey here it's a called a RAYME-data-er-base YEEEEHA!"
Additionally, after backup/restore/checkpoint/DEFRAG INDEX and memory-intensive queries or stored procedures on the SQL server, the RAM-cache becomes flushed to make room for the operation (LRU pages purged). This is why after some queries/operations the database becomes inexplicably lethargic. Many databases don't even allow the DBA to force the RDBMS to disallow LRU Primary Key main table Indexes to be flushed out of cache! We cannot control cache! This is why many databases warn against running memory-intensive queries (in violation of SQL standard), it's because they use simple caching techniques.
RAID controllers must be redesigned to minimise the load on the surviving RAID HDs during reconstruction.
UPS is old technology, the battery needs constant replacement, and very few have multiple redundant batteries and/or transistors to deal with wear and tear. Yes even a simple MOSFET transistor is not 100% reliable. Usually the only way to tell a battery is dead is your UPS fails when you need it (this happened to us when my MD was demonstrating our service live to customers, afterwards was the only time he's taken less than 9 months to sign off a purchase order on new equipment). A UPS also has a power cord to pull out when you recoil after burning your fingers on a Seagate Cheetah 15000RPM HD in the server room. A UPS also trips if you overload it, which again means the UPS fails when you most need it.
Other posts mention cosmic radiation at high altitude makes RAM fail. Last time I checked there were no Quad-Xeon Oracle databases on Concorde, although if the International Space Station were to use one this might pose a problem for non-ECC RAM. Anyway, somebody could always write a driver to do software-ECC with Reed-Solomon for RAM if it becomes necessary.
Huge databases (>500 Gigabyes) would benefit most from this as running a simple OUTER JOIN query on the biggest tables will require most of the database to be called into RAM.
- Small databases become slow due to HD latency problems if they do a lot of WRITE operations (the database is stored in RAM, the transaction log is appended to, COMMIT TRANS). This would benefit least FROM RAMdisk because a HD append operation is cheap, however it would benefit database speed in mid-backup
- Mid-size databases become HD-intensive due to aggregate queries/triggered operations over large '>RAM' datasets. For instance enforced cascading deletes where millions of tuples are being deleted cascaded to hundreds of other unindexed tables (in my job I go to the toilet whenever I run a query like this).
- Huge databases where 'Index size' > 'RAM size' - the simplest query would benefit hugely from more RAM or faster storage or RAM-storage. With current databases this would be a 10Gig Eth connection to a Terabyte RAMSan solid-state disk.
In the future, who knows, maybe a FPGA/ASIC DPU (Database Processing Unit) for INSTANT COMMIT like NVidia's GPU?I'm also confused, so I've Meta-Modded it Neutral.
Or Grid Computing in conjunction with intercorporotae colocated distributed storage
TCP/IP without incoming connections is just a selfish browsing service, like watching TV on the Internet. It's not an Internet service at all.
If we sacrifice our right to incoming TCP/IP connections then we are sacrificing our Constitutional Rights! FREEEEEDOM!
1. Go the Bank and change $1 for 5000000 Indian Rupees
Slightly exaggerated
2. Hire 1000 Indian programmers with above currency
Senior managers at HP sign off $1million invoices all the time, any expenditure less than $1million is not questioned.
3. Tell the programmers to recompile all statically-linked applications with the new libraries
Easy
4. Hire unemployed American programmer [oddtodd.com] for $20000 to translate the program from Hindi to English
Fair enough, with Indian programmers you don't need this step, I retract this particular statement.
5. Charge large corporations big $$$ for upgrading all their software
Or less $$$ than competing US corporations with US programmers paid $30000 for just one programmer.
5. Profit!!! (Really)
The CEO of Walmart is richer than Bill Gates, despite the fact the majority of his US workforce is living on charity and Government aid. Your taxpayer dollars are paying for Walmart's reduced salary, and therefore Walmart's profits. Walmart keeps its salaries low by requiring drug tests and by having patrolling managers who are under orders to stop employees talking (time theft) because "idle talk forms unions"
1. Go the Bank and change $1 for 5000000 Indian Rupees
2. Hire 1000 Indian programmers with above currency
3. Tell the programmers to recompile all statically-inked applications with the new libraries
4. Hire unemployed American programmer for $20000 to translate the program from Hindi to English
5. Charge large corporations big $$$ for upgrading all their software
5. Profit!!! (Really)
I never thought the Osprey (latest military fly-by-wire plane) would need a reset button, and that this reset button would kill people, but it does!