Using Chrome, I googled a Texas company and went to their website. I called the number listed there and got the "This call may be recorded" message before it started ringing through to the real company. When they answered, I hung up. Then I dug the real number out of my phone and called them back. I spoke to the owner of this small company, and they had no idea that it was happening. Later she reported to me that their site builder had no idea what it could be. So I did a little testing.
The only browser that changed the displayed phone number was Chrome. It was listed correctly on all the other browsers.
I then compared the html from Chrome and others, and there was an odd call replacing the static phone numbers from the original html. Not sure where this was coming from, either from the hosting site or the site generator, or google, I just mentaly flagged Chrome as suspect/venerable, and stopped using it except in static VM's (needed hangouts for work). I never did find anyone else who had noticed this behavior, or see it reported other places before. This is not exactly cogent to the OP, but I thought the audience would be the right sort of people to hear of it.
I may be crazy, but it looks to me that both booster feeds on landing were from the same Booster. Both landing motors fired at the same time, and the flames for the other Booster were visible from the top of both feeds. Either they were from the same Booster, or they were facing each other mirror image, but the overhead view of the landing pad(s) are also appear identical in the feeds, but show as not being symmetrical from from the feed views from higher elevation that show both pads and terrain.
Not saying it's faked, just that it wasn't feeds from both Boosters.
Been there, can't fit the T-shirt anymore, but this comment is pretty close to the word on the street Circa '85.
Not so sure about the commies though. I did meet and become friends with a fellow who was Russian, who was was on a team reverse engineering the ZX Spectrum in the 80's though.
Personally I was"rocking" a PDP-11 and DSM (https://en.wikipedia.org/wiki/MUMPS)
Years ago I set up a Compaq LTE LITE 4/40CX laptop to play MP3's. I think it had 4 or 8 Meg of memory, and a 1 gig hard drive. It was running Slackware, and I had to re-encode the audio stories (it was for a 10 year old relative) to mono at 128kbit, but they played just fine. It also had the docking station with a ridiculously overpowered adaptec 2940 in it to run a scanner, and a parallel port mono cam. I loved the fact that the hard drive came in an strong aluminum can that easily docked in, making multiple builds and updates as easy as shipping the new hd in the mail.
And Yes, it could play Doom. But they liked Commander Keen better.
Back in the 90's I had an office chair that made the most awful screeching noise (Freddy Kruger at a chalkboard bad). We tried everything to lubricate that chair. WD-40, dry silicon, heavy moly grease... Nothing worked, until in desperation, we tried the last thing we could come up with.
Kraft American cheese. It worked like a charm, and silenced the screech for more than a year.
I have comcast at home, and I have a 10Mbps plan, but I only get the 10Mbps infrequently. My typical speed is about 2Mbps. When I had the 6Mbps plan, the average speed was only about 1Mbps, and the best was about 4Mbps.
They don't exactly deliver what they advertised. They are more than happy to point out that they have NO guaranteed level of service, only "Typical".
As my primary usage is VPN the web page perception enhancements degrade my service and there is no opt out. All I want is an internet access provider, not a web page provider.
The 65w Athalon X2 4200 (Windsor) EE has a ACP rating of 65w, not TDP.
TDP for this processor is over 100w. WOW or COH or HL2 will easily push this processor to 65w+. The Nvidia 7600 typically pulls over 50w while in game play. It is possible that you have an underclocked 80nm GPU (Asus?). In that case, you might see 30w-45w in 3d game play. All the 7600's I've seen were 90nm GPU's. 7600GT's usually ran at 520mhz, but I did have an integrated 7650 80nm GPU at 400mhz.
What make and model is your PSU? 92% peak efficiency is very good. If your PSU is rated at 92%, that is at 50%-70% rated output. Unless you have a 300w PSU, I doubt you are getting 92%. This is especially true if it has active PFC.
Kill a watt's are a great cheap way to be aware of power usage, but they excel at showing power consumption over time, not spontaneous peak. They used to use 115 as a fixed calculation point for voltage, so higher line voltages could skewed the results. They may have fixed that by now. If you will try your test again, with your machine running the graphics benchmark of your choice, use the voltage and amperage real time readouts to get your numbers to calculate wattage. 175w for your machine "at full tilt" is not realistic, but 175w average over an hour of mixed use and some idle time is.
The HP BL680c with 4 Xeon L7345 or perhaps even E7320's and no drives in the blade should run at idle 350w-400w depending on OS. What is the power usage like at 50% load, and which CPU's do you have?
The point is the difference between hooking a couple HDD's to an old PC and buying at $700 solution could be decades of running and cooling the old PC.
My $700 number included $500 for a "couple HDD's"
so I think your point is that the cost for the rest of the system of about $150 is much more than the cost of using the old less efficient kit. Lets talk about the difference between operational cost. At the best case scenario of 85 watts for the old system and an estimated 40w for the new system, we have 45w difference.
45w increased eficiency X 24 hours X 11c/hour e- = $1.19 a day saved with the new system.
That makes the new system pay for itself in about 130 days.
130 days is somewhat less than decades.
Your Media PC is very impressive,
Super Green (no fan's) media pc "at full tilt" from the wall...
65w Athalon X2 4200 (Windsor) EE
51w generic 7600GT
20w 2 hard drives
35w "Super Green" motherboard w/sound,lan and memory (I am being very generous here, 50w is more likely)
3w keyboard,mouse
---
174w
Would you post full spec's on the system components? They seem very impressive. Do you use some kind of chimney to get a heat draw to move the heat out, or does the power supply have a fan? I am particularly interested in the make of the PSU, it appears to be 99.4% efficient!
I might have believed a Windsor 3800+ EE-SFF (35w), but not 4200 A 16 core 32GB server that doesn't pull 400w? Thats 25w per core, even Turion TL-52's are 15w/core. Just what is this server?
I am happy to discuss this further, provided you stop guessing and do some real measurements and cite hardware details.
A 25 watt P3 with 3 10 watt (idle) hard drives with no video and a super efficient 10 watt mother and memory and no fans will still use 65 watts, but with the efficiency of 70% that means 85 watts for a best case scenario for an older pc.
Since the pc does little real "work" virtually all of that is expressed as heat in your house. Fine in winter, but in summer that takes another 100 watts to cool that hot air.
An old PC full of hard drives looks cheap, but it will cost you in watts.
An old PC server can easily pull 250-400 watts continuously. And don't forget this summer, when you will have to pay twice for the waste heat.
A better solution is a VIA PC1 board, plus a couple of new drives.
Like a lot of people here, I support mission critical infrastructure, and I'm "irreplaceable"... on call 24x7x365. I always have the laptop, aircard and blackberry, at the ball game, on vacation, at home, always.
The eeepc is small enough to carry around, and big enough to use for hours at a time. Anything smaller is just too small, anything bigger reminds me that it's a ball and chain.
I have the 1g/8g model, and dual boot Xandros and XP (curse you Nortel). Improvements would be 2'nd sd slot, screen=all of lid, bluetooth and cell modem internal.
What is happening there is that the cohesive water is acting as a flexible seal around the bottom of your glass, and it is resting on a cushion of trapped air.
When the glass is put down, if the water seal forms before the glass has fully contacted the surface, the air pressure will lift the glass as it evens out the pressure on the air cushion. This will cause it to be riding on an air bearing, and slide very easily.
Usually it will only go until the water seal is broken, releasing the air pressure that forms the cushion. If the surface is moist enough, the cohesive water can renew the seal as it glides.
If you try this with a very flat, nearly level surface and a glass with a concave bottom you can get good results. Hot liquids can actually expand the air under them and suddenly lift up and slide.
Not so funny really, There were a lot of readers probably over 200K before there were accounts, and every site wanted you to have accounts to pound you with spam. Also, at first the accounts were effectively worthless, so I didn't bother. I think I did finally create an account because of some preference about something really bothersome at the top of the screen...
I think a lot of the user base did the same thing within a few months.
I still have the old asus p55t2p4 P100 that I first used to read slashdot on.
At the beginning I didn't see any point to an account. I never created an account until I needed to change the preferences. I bet thats true of a lot of the 200K people.
If the ad's ever get ahead of my adblocking I'll probably pay for one.
Because healthy, genetically perfect people can catch fatal diseases (antibiotic resistant tuberculosis anyone) from the bum on the corner who can't afford insurance. You have a vested interest in that bum's good health.
Should only people who's house is on fire pay taxes to support the firemen? Maybe it should be everyone who doesn't want to burn to death.
The Public's health is everyones concern, that should be obvious.
I hate to point this out to you, but when you have something happen that causes that $20,000 bill, you now are one of the unhealthy people that people are now "good-heartedly subsidizing". That's your INSURANCE.
You wouldn't want to catch pneumonia (or antibiotic resistant tuberculosis) from that bum on the corner who can't afford 'healthy/employed people" insurance. He probably could infect 2000 people a day... You have a vested interest in that bum's good health. "other equally healthy people" aren't necessarily healthy tomorrow, and no amount of good genetics or low BMI will prevent that.
>> It's good for a nice surprise too: I had several people at my desk that were trying to move the mouse pointer by moving the whole device around:)
But the nice thing is it is never a surprise to you!
It's always exactly where you left it, and it never moves. When keyboarding (most of the time for me) I often need to use the mouse for a quick operation, for this the Marble mouse is perfect, as you can reach over to the same position each time with no fumbling.
My Marble mouse has 4 buttons, with the standard right/left as large buttons on each side, with the other two as small cutouts at the top of each left/right button.
In UT / HL / ID I map these instead of the scroll. You might try the reverse mouse option as well, I do that when I have to use mice sometimes, depending on the game. I do wish that they would add a scrolls on each side below the button, where you could really use it. the scroll on top is wasted for me, as I use the three middle fingers to roll the ball, and thumb/pinky to click. Center/top scroll buttons seem to make the back of my hand hurt after a while.
As a long time Linux user, I couldn't get by without a good easy to use center click button, and the center click/scroll button is poor for this. I use the Marble mouses small inboard buttons for that as well. They are easy to get to but not in the way.
If I use the term "Scroll Buttons" one more time, I'm gonna have to make a StrongBad reference.
You are right about Linux systems and Swap, but I still allocate at least 20G of swap to my AMD/Intel systems.
Think of it as your window of time to find and kill that runaway process. and Hey...the DB guys just added in annother environment without letting anyone know.
_Using_ no swap is the ideal for performance,
but
_Having_ swap is nice if you like to sleep.
That plus, if the users never suffer, they take you for granted.
Well then - we're in the same ballpark. However, I'm not a pedantic asshole, err, bastard.
My friends call me asshole, feel free. I do think you are being pedantic though.
So, let's take you to school:
Always happy to learn.
Lets see...
"--- very fast devices at both ends. This generally means RAID 0, 5, 6 or some combination thereof like 50 or 60.
Raid 10 is more for redundancy in case a disk dies than performance"
Nope, no speed by sequence there.
Let's see - I dropped RAID 3,4,30, and 40. Guess what, they're generally not used anymore. (Note - for your pedanticness, in general, ie, normally, ie, not specific use cases,
they're not used anymore. I just hit someone's comma's in a sentence limit.) This is just a natural sequencing of numbers there.
3's are still in use, but they are for people stuck in particular hardware. They suck.
"Second, RAID 5 with 8 disks is certainly faster than RAID 0 or RAID 10 with less than 6 equivalent disks"
If this is true of your setup, then there is something deeply wrong with your controller setup.
Either you have a controller with a fast cpu and lots of cache + really really poor disks, or it doesn't really do raid 10.
Why yes, I do have controllers with buttloads of cache AND a fast CPU. Why would I run enterprise systems on the cheap?
You missed the point. for this to be true, your disks have to be truly awfull, way way out of scale to the controller.
"I have owned about 8 different types, although none from the last 2-3 years - ie, IDE/SATA)"
Ok, now I'm getting the picture, you are a Hobbyist.
Not exactly, I dropped out of enterprise system hardware support about 5 years ago. Since then, I checked out the initial set of IDE RAID cards, and discovered they still
didn't overcome issue #1 with IDE - namely, multiple I/O. This means IDE of any flavor sucks for performance. They're great as mass store devices though.
Lots of enterprise class people are running fibre attached sata nowadays. Some business just can't see paying $1500 more for a drive to get that extra 15%
"RAID 6 would require a minimum of 5 drives"
Sorry, its only 4
That's true. Raid 5 only requires 3. Raid 5 with 3 drives, or raid 6 with 4, will suck eggs and isn't recommended. It's kinda like running windows with a minimum configuration -
it'll work, but it will suck. Even with 4 and 5 drives respectively, it won't be great.
Granted. Raid 5 sucks for writes, is passable for reads, and career ending on the rebuild. And if it looses 2 drives, it's garbage Over and over I've seen it. One drive fails and the hot spare rebuild kills another drive, leaving you praying you beat the 60% restore failure odds.
Raid 10 has no parity to rebuld. The hot spare rebuild is just a re-mirror, and only affects ONE other drive in the array, plus, you can loose up to half the drives at a time. Just having the quick rebuild without significant storage performance degradation is worth the price of the other drives. With raid 10, you find out you lost a drive when the log reports send you email. With raid 5 you find out when the pager and phone go off, and your explaining to the VP that the system w
Slashdot Mirror
Using Chrome, I googled a Texas company and went to their website. I called the number listed there and got the "This call may be recorded" message before it started ringing through to the real company. When they answered, I hung up.
Then I dug the real number out of my phone and called them back. I spoke to the owner of this small company, and they had no idea that it was happening. Later she reported to me that their site builder had no idea what it could be.
So I did a little testing.
The only browser that changed the displayed phone number was Chrome. It was listed correctly on all the other browsers.
I then compared the html from Chrome and others, and there was an odd call replacing the static phone numbers from the original html.
Not sure where this was coming from, either from the hosting site or the site generator, or google, I just mentaly flagged Chrome as suspect/venerable, and stopped using it except in static VM's (needed hangouts for work). I never did find anyone else who had noticed this behavior, or see it reported other places before.
This is not exactly cogent to the OP, but I thought the audience would be the right sort of people to hear of it.
Be Canadian proud anyway, Elon Musk is a Canadian citizen (and American, and South African)
I may be crazy, but it looks to me that both booster feeds on landing were from the same Booster. Both landing motors fired at the same time, and the flames for the other Booster were visible from the top of both feeds. Either they were from the same Booster, or they were facing each other mirror image, but the overhead view of the landing pad(s) are also appear identical in the feeds, but show as not being symmetrical from from the feed views from higher elevation that show both pads and terrain.
Not saying it's faked, just that it wasn't feeds from both Boosters.
Been there, can't fit the T-shirt anymore, but this comment is pretty close to the word on the street Circa '85.
Not so sure about the commies though. I did meet and become friends with a fellow who was Russian, who was was on a team reverse engineering the ZX Spectrum in the 80's though.
Personally I was"rocking" a PDP-11 and DSM (https://en.wikipedia.org/wiki/MUMPS)
Years ago I set up a Compaq LTE LITE 4/40CX laptop to play MP3's. I think it had 4 or 8 Meg of memory, and a 1 gig hard drive. It was running Slackware, and I had to re-encode the audio stories (it was for a 10 year old relative) to mono at 128kbit, but they played just fine. It also had the docking station with a ridiculously overpowered adaptec 2940 in it to run a scanner, and a parallel port mono cam. I loved the fact that the hard drive came in an strong aluminum can that easily docked in, making multiple builds and updates as easy as shipping the new hd in the mail.
And Yes, it could play Doom. But they liked Commander Keen better.
Back in the 90's I had an office chair that made the most awful screeching noise (Freddy Kruger at a chalkboard bad). We tried everything to lubricate that chair. WD-40, dry silicon, heavy moly grease... Nothing worked, until in desperation, we tried the last thing we could come up with.
Kraft American cheese. It worked like a charm, and silenced the screech for more than a year.
I have comcast at home, and I have a 10Mbps plan, but I only get the 10Mbps infrequently. My typical speed is about 2Mbps. When I had the 6Mbps plan, the average speed was only about 1Mbps, and the best was about 4Mbps.
They don't exactly deliver what they advertised. They are more than happy to point out that they have NO guaranteed level of service, only "Typical".
As my primary usage is VPN the web page perception enhancements degrade my service and there is no opt out.
All I want is an internet access provider, not a web page provider.
The 65w Athalon X2 4200 (Windsor) EE has a ACP rating of 65w, not TDP.
TDP for this processor is over 100w.
WOW or COH or HL2 will easily push this processor to 65w+.
The Nvidia 7600 typically pulls over 50w while in game play. It is possible that you have an underclocked 80nm GPU (Asus?). In that case, you might see 30w-45w in 3d game play. All the 7600's I've seen were 90nm GPU's. 7600GT's usually ran at 520mhz, but I did have an integrated 7650 80nm GPU at 400mhz.
What make and model is your PSU? 92% peak efficiency is very good.
If your PSU is rated at 92%, that is at 50%-70% rated output. Unless you have a 300w PSU, I doubt you are getting 92%. This is especially true if it has active PFC.
Kill a watt's are a great cheap way to be aware of power usage, but they excel at showing power consumption over time, not spontaneous peak. They used to use 115 as a fixed calculation point for voltage, so higher line voltages could skewed the results. They may have fixed that by now. If you will try your test again, with your machine running the graphics benchmark of your choice, use the voltage and amperage real time readouts to get your numbers to calculate wattage. 175w for your machine "at full tilt" is not realistic, but 175w average over an hour of mixed use and some idle time is.
The HP BL680c with 4 Xeon L7345 or perhaps even E7320's and no drives in the blade should run at idle 350w-400w depending on OS. What is the power usage like at 50% load, and which CPU's do you have?
The point is the difference between hooking a couple HDD's to an old PC and buying at $700 solution could be decades of running and cooling the old PC.
My $700 number included $500 for a "couple HDD's" so I think your point is that the cost for the rest of the system of about $150 is much more than the cost of using the old less efficient kit.
Lets talk about the difference between operational cost. At the best case scenario of 85 watts for the old system and an estimated 40w for the new system, we have 45w difference.
45w increased eficiency X 24 hours X 11c/hour e- = $1.19 a day saved with the new system.
That makes the new system pay for itself in about 130 days.
130 days is somewhat less than decades.
Your Media PC is very impressive,
Super Green (no fan's) media pc "at full tilt" from the wall...
65w Athalon X2 4200 (Windsor) EE
51w generic 7600GT
20w 2 hard drives
35w "Super Green" motherboard w/sound,lan and memory (I am being very generous here, 50w is more likely)
3w keyboard,mouse
---
174w
Would you post full spec's on the system components? They seem very impressive. Do you use some kind of chimney to get a heat draw to move the heat out, or does the power supply have a fan? I am particularly interested in the make of the PSU, it appears to be 99.4% efficient!
I might have believed a Windsor 3800+ EE-SFF (35w), but not 4200
A 16 core 32GB server that doesn't pull 400w? Thats 25w per core, even Turion TL-52's are 15w/core. Just what is this server?
I am happy to discuss this further, provided you stop guessing and do some real measurements and cite hardware details.
Or perhaps you just need to get in the last word.
I get my numbers from an actual measurements. Where do you get yours? If you don't have an meter, play with this for a bit.
if your interested in how it works you can read this or a less technical WIKI
Older Processors used lots of power too, and the old power supplies were usually less than 70% efficient
A 25 watt P3 with 3 10 watt (idle) hard drives with no video and a super efficient 10 watt mother and memory and no fans will still use 65 watts, but with the efficiency of 70% that means 85 watts for a best case scenario for an older pc.
Since the pc does little real "work" virtually all of that is expressed as heat in your house. Fine in winter, but in summer that takes another 100 watts to cool that hot air.
Here is a nice article from 2000 that has real measured usage of these now vintage machines-----
and another from 2004>
But most likely these are going to be machines from the last 2 years that can't play the new games, and will likely use far more power.
For reference, my SLI game setup (AMD64 5600 X2, 2 Raptors, 2 Asus 6600 GTS's) pulls 520 real watts playing half life.
I can only afford to play in winter. :P
An old PC full of hard drives looks cheap, but it will cost you in watts. An old PC server can easily pull 250-400 watts continuously. And don't forget this summer, when you will have to pay twice for the waste heat.
A better solution is a VIA PC1 board, plus a couple of new drives.
The "$60 PC 1" will only pull 20 watts at max. Combine this with 2 "$250 terabyte drives" mirrored, and a small low wattage "$35 case" and the "(Free) Linux" of your choice,
You will have a reliable Terabyte server for less than $700, that only pulls as much power as a small appliance bulb.
Like a lot of people here, I support mission critical infrastructure, and I'm "irreplaceable"... on call 24x7x365.
I always have the laptop, aircard and blackberry, at the ball game, on vacation, at home, always.
The eeepc is small enough to carry around, and big enough to use for hours at a time.
Anything smaller is just too small, anything bigger reminds me that it's a ball and chain.
I have the 1g/8g model, and dual boot Xandros and XP (curse you Nortel).
Improvements would be 2'nd sd slot, screen=all of lid, bluetooth and cell modem internal.
I'm typing this now on my eeepc in my car.
What is happening there is that the cohesive water is acting as a flexible seal around the bottom of your glass, and it is resting on a cushion of trapped air.
When the glass is put down, if the water seal forms before the glass has fully contacted the surface, the air pressure will lift the glass as it evens out the pressure on the air cushion. This will cause it to be riding on an air bearing, and slide very easily.
Usually it will only go until the water seal is broken, releasing the air pressure that forms the cushion. If the surface is moist enough, the cohesive water can renew the seal as it glides.
If you try this with a very flat, nearly level surface and a glass with a concave bottom you can get good results.
Hot liquids can actually expand the air under them and suddenly lift up and slide.
Not so funny really, There were a lot of readers probably over 200K before there were accounts, and every site wanted you to have accounts to pound you with spam. Also, at first the accounts were effectively worthless, so I didn't bother. I think I did finally create an account because of some preference about something really bothersome at the top of the screen...
I think a lot of the user base did the same thing within a few months.
I still have the old asus p55t2p4 P100 that I first used to read slashdot on.
BINGO!!!
At the beginning I didn't see any point to an account. I never created an account until I needed to change the preferences. I bet thats true of a lot of the 200K people.
If the ad's ever get ahead of my adblocking I'll probably pay for one.
Yeah, I've had nothing but trouble from this consulting group.
http://www.newtechusa.com/ppi/talent.asp
One of the hazards of browsing at 4+ moderation. I didn't see the disclaimers until too late.
I bet the "insightful" galls you a bit.
It did me.
Because healthy, genetically perfect people can catch fatal diseases (antibiotic resistant tuberculosis anyone) from the bum on the corner who can't afford insurance.
You have a vested interest in that bum's good health.
Should only people who's house is on fire pay taxes to support the firemen?
Maybe it should be everyone who doesn't want to burn to death.
The Public's health is everyones concern, that should be obvious.
Don't forget people who:
live in big dirty cities
drive fast cars
have tattoo's
go to bars
have a life
I hate to point this out to you, but when you have something happen that causes that $20,000 bill, you now are one of the unhealthy people that people are now "good-heartedly subsidizing". That's your INSURANCE.
You wouldn't want to catch pneumonia (or antibiotic resistant tuberculosis) from that bum on the corner who can't afford 'healthy/employed people" insurance. He probably could infect 2000 people a day...
You have a vested interest in that bum's good health. "other equally healthy people" aren't necessarily healthy tomorrow, and no amount of good genetics or low BMI will prevent that.
The Public's health is everyones concern.
>> It's good for a nice surprise too: I had several people at my desk that were trying to move the mouse pointer by moving the whole device around :)
But the nice thing is it is never a surprise to you!
It's always exactly where you left it, and it never moves. When keyboarding (most of the time for me) I often need to use the mouse for a quick operation, for this the Marble mouse is perfect, as you can reach over to the same position each time with no fumbling.
My Marble mouse has 4 buttons, with the standard right/left as large buttons on each side, with the other two as small cutouts at the top of each left/right button.
In UT / HL / ID I map these instead of the scroll. You might try the reverse mouse option as well, I do that when I have to use mice sometimes, depending on the game.
I do wish that they would add a scrolls on each side below the button, where you could really use it. the scroll on top is wasted for me, as I use the three middle fingers to roll the ball, and thumb/pinky to click. Center/top scroll buttons seem to make the back of my hand hurt after a while.
As a long time Linux user, I couldn't get by without a good easy to use center click button, and the center click/scroll button is poor for this. I use the Marble mouses small inboard buttons for that as well. They are easy to get to but not in the way.
If I use the term "Scroll Buttons" one more time, I'm gonna have to make a StrongBad reference.
I have been using this same exact mouse for the last 6 years at work. I have 4 others for the utility and games machines at home.
When they make a better mouse, I'll switch in a heartbeat.
It hasn't happened yet.
off-topic
I like your sig. mine is similar, but the inverse.
It used to be better, but I had to trim it when they started counting the formatting char.
If you are the Doug Moen I think you may be, thanks for all the hard work.
You are right about Linux systems and Swap, but I still allocate at least 20G of swap to my AMD/Intel systems.
Think of it as your window of time to find and kill that runaway process.
and
Hey...the DB guys just added in annother environment without letting anyone know.
_Using_ no swap is the ideal for performance,
but
_Having_ swap is nice if you like to sleep.
That plus, if the users never suffer, they take you for granted.
Well then - we're in the same ballpark. However, I'm not a pedantic asshole, err, bastard.
My friends call me asshole, feel free. I do think you are being pedantic though.
So, let's take you to school:
Always happy to learn.
Lets see...
"--- very fast devices at both ends. This generally means RAID 0, 5, 6 or some combination thereof like 50 or 60.
Raid 10 is more for redundancy in case a disk dies than performance"
Nope, no speed by sequence there.
Let's see - I dropped RAID 3,4,30, and 40. Guess what, they're generally not used anymore. (Note - for your pedanticness, in general, ie, normally, ie, not specific use cases,
they're not used anymore. I just hit someone's comma's in a sentence limit.) This is just a natural sequencing of numbers there.
3's are still in use, but they are for people stuck in particular hardware. They suck.
"Second, RAID 5 with 8 disks is certainly faster than RAID 0 or RAID 10 with less than 6 equivalent disks"
If this is true of your setup, then there is something deeply wrong with your controller setup.
Either you have a controller with a fast cpu and lots of cache + really really poor disks, or it doesn't really do raid 10.
Why yes, I do have controllers with buttloads of cache AND a fast CPU. Why would I run enterprise systems on the cheap?
You missed the point. for this to be true, your disks have to be truly awfull, way way out of scale to the controller.
"I have owned about 8 different types, although none from the last 2-3 years - ie, IDE/SATA)"
Ok, now I'm getting the picture, you are a Hobbyist.
Not exactly, I dropped out of enterprise system hardware support about 5 years ago. Since then, I checked out the initial set of IDE RAID cards, and discovered they still
didn't overcome issue #1 with IDE - namely, multiple I/O. This means IDE of any flavor sucks for performance. They're great as mass store devices though.
Getging better all the time.
http://www.redbooks.ibm.com/Redbooks.nsf/RedbookAb stracts/tips0458.html?Open
Lots of enterprise class people are running fibre attached sata nowadays. Some business just can't see paying $1500 more for a drive to get that extra 15%
"RAID 6 would require a minimum of 5 drives"
Sorry, its only 4
That's true. Raid 5 only requires 3. Raid 5 with 3 drives, or raid 6 with 4, will suck eggs and isn't recommended. It's kinda like running windows with a minimum configuration -
it'll work, but it will suck. Even with 4 and 5 drives respectively, it won't be great.
Granted.
Raid 5 sucks for writes, is passable for reads, and career ending on the rebuild. And if it looses 2 drives, it's garbage
Over and over I've seen it. One drive fails and the hot spare rebuild kills another drive, leaving you praying you beat the 60% restore failure odds.
Raid 10 has no parity to rebuld. The hot spare rebuild is just a re-mirror, and only affects ONE other drive in the array, plus, you can loose up to half the drives at a time.
Just having the quick rebuild without significant storage performance degradation is worth the price of the other drives.
With raid 10, you find out you lost a drive when the log reports send you email.
With raid 5 you find out when the pager and phone go off, and your explaining to the VP that the system w