Each step adds cost and it is still harder to train a monkey to do any of those things than "look for the red blinking light, pull the handle and slide the drive out. Push in the new drive and walk away
They discussed this was being done in software, and while that is great, you still will have a slowdown with RAID 6. Will it be unbearable? surely not. is it RAID 10? Not a chance, but I think they care more about storage than performance for this. Also, with RAID 6 in software with mdadm they did not discuss the raid stripe configuration and whether it was left or right sync or async, etc.
The hardest part will be identifying the bad drives. That is ANOTHER feature that you pay for on expensive disk systems. The controllers will alert you to where the failed drive is, as well as often times alerting the manufacturer of the failure. There have been times I have been called by a vendor to let me know a part and on site engineer was being dispatched for a failure my users were not even aware of yet due to it being off hours (and ops were asleep at the wheel).
You will more than likely NOT have to take a node offline. The design looks like they place the drives into slip down hot plug enclosures. Most rack mounted hardware is on rails, not screwed to the rack. You roll the rack out, log in, fail the drive that is bad, remove it, hot plug another drive and add it to the array. You are now done.
They went RAID 6, even though it is slow as shit, for the added failsafe mechanisms.
in many cases for these applications, MORE RAM is better than a faster RAM. Also, the $2000 for 2 CPUs is pittance compared to the overall cost of running those machines. You fail at math as well. You save money with a similar setup going for these CPUs. The savings on RAM are on TOP of that. These are not gaming machines. They are usually more concerned with concurrency than with sheer clock value. This is why many servers do NOT have a single fast core.
The FAIL is yours. You assume that the 6 core offerings from other manufacturers are free? The issue is that the savings in populating the RAM is STILL a $1000 net savings.
Ad hominem attacks aside, I doubt that entirely. The development community tends to be swift enough to change those technologies, or were people using Python 10 years ago or Ruby 7 years ago? There are lots of people who complain "OMG, my $FAV_APP is not supported on Linux" and with a cross platform technology, regardless of who originated it, there chances of seeing it available to them increase.
Also, if somebody were say on Windows Server 2003 and said "Shit, I don't want to pay to upgrade 4000 servers plus support for this crap to 2008, let me look at going with Red Hat or SUSE, but oh no, we have all this.NET code we wrote..." They don't have to pucker up to the computer masters at Microsoft and shell out for this upgrade. They can move their entire code base to another OS platform..NET does not need broader appeal. People already embraced it on the MS front. There are thousands of openings for.NET programmers. There are still plenty of openings for Java, Python, PHP, Perl, C, C++, COBOL programmers out there. What this does is actually give Linux users some leverage.
Don't fool yourselves. Cops are not great with their guns. Most wear them on their belt along with a spare magazine or two and a flashlight and cuffs. They are rarely used unless the yearly qualifications are due and even then, maybe 200 rounds are expended. Last year alone, I, not a cop, shot over 2500 rounds of 9mm and another 1500rds of.22 at the same targets they use. The idea of seeing that they need to put ~40 out of 50 shots anywhere in a man sized target at 20yds is not inspiring of confidence. The target is not moving and not shooting back.
Never let facts get in the way of their ramblings, especially Roy. He foams at the mouth but never actually got to the reason WHY any deal was developed. Novell tried to embrace interoperability and was told that they should join as the same deal was given to Red Hat, et al, and they thought "OK, sure, lets make this work and protect our customers."
Novell contributes code to the same thing the boycottnovell mouth breathers use every day like KDE, Gnome, SAMBA and plenty of others, along with being part of the Open Invention Network using their patent portfolio as a shield. They are, at least for now, the good guys. The future may change. Also, while some may hate Mono, it opens the door to running.NET apps on Linux so its a win in a way.
How do we have different companies for other items like phone service? Wait.......I know, we don't. I mean it is not like you can call Verizon and say "Bye, going to AT&T."
The government's strong regulatory powers GOT US HERE. If Comcast did not have legal monopolies on cable where it operates, competition would force this to change. The government has your back the same way that any bunch of thieves do. As long as it suits them, they will.
Hyperic makes a really good java based app/server/network monitoring and management tool. Their stuff is really good. However, as their offering is "FOSS" but crippled, I have looked more towards Pandora FMS. They make Nagios look like........nagios, for complexity.
I doubt it. Just today, I found out that some dude named Sam Clemens was ripping off Rush, and using an assumed name to make money off their Tom Sawyer song. What a douche......
I'm glad that you can assume a performance margin without knowing the workload or the application. Please, enlighten us with the performance of ZFS using Oracle or another database...Sure it can be fast, but please, in detail, explain the tunables that need to be set to achieve this performance and what kind of issues you may have with fsync and such, especially when dealing with SAN storage with external battery backed cache..... I am curious..... (and yet know the answers).
Ok, go talk to Falconstor, EMC, HP, IBM, SUN/STK, etc and tell them that virtual tape libraries are shit and that nobody should use them. Perhaps you should say "Single disk backup solutions are bad" but not that hard drives are not backup media. When you dedupe your backups and sync across continents, do you want to read 40 800GB tapes per week? I sure as shit don't. Deduping tape is horrendous. Deduping large arrays is relatively painless and when you can make the bar codes match the virtual tape device and the off site tape that is cut after the backup to disk (Yes, a Tiered approach) you make life so much easier.
This is insightful? WTF? Is there some basement bereft of an idiot who thinks flashing his xbox makes him a sysadmin? Roll-outs where I am take months to implement as we have to coordinate with multiple countries. Sure, in a smaller environment, you want to test against applications, user auth, speed/performance testing, give time to account for problems that may need a fix or patch, etc, draw up a patching schedule and gain competency with the staff.
if it was a default FS on the latest version of the $OS_Shipped_On_95_Percent_Of_Desktops and had this bug, sure. If it is a relatively new and untested file system on an OS with choices of stable FS like Reiser, Ext2/3, JFS, XFS, OCFS2, etc, then no, not as big a deal....
Slashdot Mirror
Each step adds cost and it is still harder to train a monkey to do any of those things than "look for the red blinking light, pull the handle and slide the drive out. Push in the new drive and walk away
Yes, and the also should support some RAID features, I believe.
They discussed this was being done in software, and while that is great, you still will have a slowdown with RAID 6. Will it be unbearable? surely not. is it RAID 10? Not a chance, but I think they care more about storage than performance for this. Also, with RAID 6 in software with mdadm they did not discuss the raid stripe configuration and whether it was left or right sync or async, etc.
The hardest part will be identifying the bad drives. That is ANOTHER feature that you pay for on expensive disk systems. The controllers will alert you to where the failed drive is, as well as often times alerting the manufacturer of the failure. There have been times I have been called by a vendor to let me know a part and on site engineer was being dispatched for a failure my users were not even aware of yet due to it being off hours (and ops were asleep at the wheel).
You will more than likely NOT have to take a node offline. The design looks like they place the drives into slip down hot plug enclosures. Most rack mounted hardware is on rails, not screwed to the rack. You roll the rack out, log in, fail the drive that is bad, remove it, hot plug another drive and add it to the array. You are now done.
They went RAID 6, even though it is slow as shit, for the added failsafe mechanisms.
in many cases for these applications, MORE RAM is better than a faster RAM. Also, the $2000 for 2 CPUs is pittance compared to the overall cost of running those machines. You fail at math as well. You save money with a similar setup going for these CPUs. The savings on RAM are on TOP of that. These are not gaming machines. They are usually more concerned with concurrency than with sheer clock value. This is why many servers do NOT have a single fast core.
The FAIL is yours. You assume that the 6 core offerings from other manufacturers are free? The issue is that the savings in populating the RAM is STILL a $1000 net savings.
Ad hominem attacks aside, I doubt that entirely. The development community tends to be swift enough to change those technologies, or were people using Python 10 years ago or Ruby 7 years ago? There are lots of people who complain "OMG, my $FAV_APP is not supported on Linux" and with a cross platform technology, regardless of who originated it, there chances of seeing it available to them increase. Also, if somebody were say on Windows Server 2003 and said "Shit, I don't want to pay to upgrade 4000 servers plus support for this crap to 2008, let me look at going with Red Hat or SUSE, but oh no, we have all this .NET code we wrote..." They don't have to pucker up to the computer masters at Microsoft and shell out for this upgrade. They can move their entire code base to another OS platform. .NET does not need broader appeal. People already embraced it on the MS front. There are thousands of openings for .NET programmers. There are still plenty of openings for Java, Python, PHP, Perl, C, C++, COBOL programmers out there. What this does is actually give Linux users some leverage.
Don't fool yourselves. Cops are not great with their guns. Most wear them on their belt along with a spare magazine or two and a flashlight and cuffs. They are rarely used unless the yearly qualifications are due and even then, maybe 200 rounds are expended. Last year alone, I, not a cop, shot over 2500 rounds of 9mm and another 1500rds of .22 at the same targets they use. The idea of seeing that they need to put ~40 out of 50 shots anywhere in a man sized target at 20yds is not inspiring of confidence. The target is not moving and not shooting back.
Never let facts get in the way of their ramblings, especially Roy. He foams at the mouth but never actually got to the reason WHY any deal was developed. Novell tried to embrace interoperability and was told that they should join as the same deal was given to Red Hat, et al, and they thought "OK, sure, lets make this work and protect our customers."
Novell contributes code to the same thing the boycottnovell mouth breathers use every day like KDE, Gnome, SAMBA and plenty of others, along with being part of the Open Invention Network using their patent portfolio as a shield. They are, at least for now, the good guys. The future may change. Also, while some may hate Mono, it opens the door to running .NET apps on Linux so its a win in a way.
How do we have different companies for other items like phone service? Wait.......I know, we don't. I mean it is not like you can call Verizon and say "Bye, going to AT&T."
The government's strong regulatory powers GOT US HERE. If Comcast did not have legal monopolies on cable where it operates, competition would force this to change. The government has your back the same way that any bunch of thieves do. As long as it suits them, they will.
Send sweaters
HypericHQ would be good
PandoraFMS is another option.
ZenOSS and Zabbix are popular too
asterisk, moron. I'd have even given you "splat" but you'd misspell that too.
God if I didn't have stupid users ask me shit like "It says Install disk 2 to continue and I keep hitting Continue but nothing is happening" ......
Hyperic makes a really good java based app/server/network monitoring and management tool. Their stuff is really good. However, as their offering is "FOSS" but crippled, I have looked more towards Pandora FMS. They make Nagios look like........nagios, for complexity.
I doubt it. Just today, I found out that some dude named Sam Clemens was ripping off Rush, and using an assumed name to make money off their Tom Sawyer song. What a douche......
What happens when that disk goes loose? Do you have to tighten her back up or just ask for a daddy stitch?
I'm glad that you can assume a performance margin without knowing the workload or the application. Please, enlighten us with the performance of ZFS using Oracle or another database...Sure it can be fast, but please, in detail, explain the tunables that need to be set to achieve this performance and what kind of issues you may have with fsync and such, especially when dealing with SAN storage with external battery backed cache..... I am curious..... (and yet know the answers).
http://reductivelabs.com/products/ That will be $4000 in consulting fees, thanks.
FAT chance.........
Ok, go talk to Falconstor, EMC, HP, IBM, SUN/STK, etc and tell them that virtual tape libraries are shit and that nobody should use them. Perhaps you should say "Single disk backup solutions are bad" but not that hard drives are not backup media. When you dedupe your backups and sync across continents, do you want to read 40 800GB tapes per week? I sure as shit don't. Deduping tape is horrendous. Deduping large arrays is relatively painless and when you can make the bar codes match the virtual tape device and the off site tape that is cut after the backup to disk (Yes, a Tiered approach) you make life so much easier.
Two weeks is fucking optimistic.
if it was a default FS on the latest version of the $OS_Shipped_On_95_Percent_Of_Desktops and had this bug, sure. If it is a relatively new and untested file system on an OS with choices of stable FS like Reiser, Ext2/3, JFS, XFS, OCFS2, etc, then no, not as big a deal....