I was just laid off from Xiotech,(with about 100 others) one of the smaller SAN vendors..
But we call the Centerra a "data jail". It's like the roach motel..
Data checks in, but it don't check out. It can't scale beyond a 42U rack enclosure. It's a bunch of little servers striped together to form a big NAS with a metedata controller in the middle.
Just bashing the 800lb gorilla... OTOH, If you're hiring, I'm willing to tell you your products could rule the world!
And I AM a geek. I find it to be an excellent resource for dealing with people.
I mean all kind of people.
It was recommended to me by my mentor, and I can see him apply it almost daily.
Amazing resource.
Don't Confuse
on
SATA vs ATA?
·
· Score: 4, Insightful
The connection technology with the drive / spindle quality.
(P)ATA and SATA are connection technologies. They have their individual benefits and drawbacks (cost, reliability, speed)
The real factors to consider are the details of the drives themselves - vibration dampening, bearing and motor quality, MTBF.
It used to be rather simple to guess what quality of drive you were buying. If it was 146GB or less (73GB, 36GB), and rotational speed was 10K or 15K, it was either SCSI or FC, and an "enterprise" class drive, rated in Mean Time Between Failure.
Good drive, high quality, expect it to last several years, spinning 24 hours a day, sustaining high read and write activity during production and backup hours.
If the drive was larger (200GB+) and slower (7200 RPM), typically an ATA drive, maybe low end SCSI.
Then it was, at best, a workstation class drive, rated in "Contact Start Stops", meaning how many spin-ups and shutdowns the drive should survive. Not meant to run 24 hours a day, and run under heavy load except for short periods.
The lines are beginning to blur with 300 - 500 GB drives with FC drive attachment. Those drives are meant for archiving and reference data. Not production databases and such.
In my personal experience, the 3Ware products are worth the premium.
Pick your attachment technology as appropriate.
Best of Luck, Patrick (slineyp at hotmail dot com)
From what I read, the XSan software is first and foremost a distributed file system for shared volumes from the Xserve RAID. If you look at the applications, it's about multiple servers or workstations with concurrent access to a single volume - distributed file locking.
Great stuff for the stated purpose, can't wait to get my hands on it!
Hiding the complexity of RAID is the domain of storage 'virtualization' solutions. The ones that let you mix and match raid types across any number of spindles you throw at it.
<Shameless_Plug> My product, the XIOtech Magnitude does that. Take up to 126 spindles, create RAID 0, 1, 5, 10 volumes and give 'em to your servers. Boot off 'em, mirror 'em, copy 'em. Stick 'em in your ear! </Shameless_Plug>
direct flames or questions to slineyp at hotmail dot com
When that slow 250GB ATA class drive is dead, and while its fellow drives are chugging their little hearts out (and probably maxing out that 3ware controller), how long will it take to rebuild your array?
Have you tested how long it takes? Probably better than 24 hours if your system is moderately loaded.
Guess what you have now? The marvelous opportunity for a CASCADING FAILURE!
That's right kids! Because you just had a drive fail, and all the other drives are doing double the work to rebuild from parity data, you have a higher chance of getting a second drive failure.
Consider that you bought all of the drives in that array at the same time. They've all been running the same amount of time. What if there was a minor manufacturing defect that caused that First drive to fail? How soon before it takes out the other 4?
A 'resume generating event' waiting to happen.
Best of luck.. and I agree with the comment upthread. SATA drives are for Workstations. Maybe for storing what we call 'reference data'. Not much more.
There's a few choice terms in the industry- 'Economy Enterprise' 'Garbage RAID' 'Ghetto SAN'
It takes a little setup, and a little cash, but I believe it makes for the best lab setup possible. I've designed several setups for customers like this:
- As many servers as you need (8, for example) Buy 8 identical servers - explained later.
- A Fibre Channel HBA for each host.
- A Fibre Channel disk array. Preferrably an old XIOtech Magnitude, available on eBay now for $14K - $20K, new from around $60K
- Connect up to 8 hosts directly to the Magnitude, assign a boot volume to the host from the Mag's management interface. Tell the host HBA to boot from the attached device
- Load the OS of your choice (most now support FC HBAs as boot devices). You can load multiple OS on each box, and swap logical hosts between physical boxes if need be.
- Lather, Rinse, Repeat.
email me with any additional questions- slineyp at hotmail dot com
Lets take this offlist, I've got a question about the EVAs and the Auto RAID function
slineyp at hotmail dot com - thanks
but you are right about spreading your data and I/Os around.
The reason customers pay a large fortune for EMC / HDS systems is that they're designed to get ungodly performance (cache based RAID) and that near-mythical "five 9s" availability (redundant disks, cache, backplane, power, etc). It seems to be a disagreement in design circles about using modular arrays like the EVAs as nearly a "Field Replacable Unit" versus using monolithic HA designs like the EMC Symm or HDS lightning.
Here Here! I hope JNI folds soon. I'm so glad QLogic is kicking their ass in the Sun market. I'm also recommending QLogic -E(MC) certified cards to my EMC customers.
We're not held captive to that crap anymore!
The Emulex LPUTIL has to be the single least intuitive POS configuration utility I've ever seen in a Fibre Channel product (and that's saying a lot considering the Brocade mgmt interface)
Hit the nail on the head, it's likely the encryption device, that sits between the radio input and the transmitter proper.
The reason they want it back is that is IS a CCI, and therefore by beuracratic law and common sense, must be recovered and accounted for.
As to the key not being intact, odds are the device was in use when crew perished. Likely they didn't zeroize (official term, no shit) the key. If the unit is intact, so is the key. Fortunately, the key storage space is "tamper proof" that would self destruct the storage area on any attempt to crack it open.
If you have 3.5 TB AND the capital to get an autoloader, then by all means!! Depending on your environment (you made no mention of what OS's, where the capacity is concentrated, etc.) you should be able to call one of the b/u software vendors directly (CA, Veritas, Legato, Bakbone, CommVault, etc) or call one of their channel partners and start sniffing for a deal. I work in this sector and man, everyone is slugging it out and dropping their prices to bottom just to win business.
Also consider some storage management software like SRM- It will let you know EXACTLY what you are backing up, who's hogging space (by user and group) and where your capacity is allocated (an mis-allocated)
The short version is this: With some SRM tools and a good automated backup system, you WILL reduce your workload. I know, I set this stuff up for a living.
Disclaimer: I am a tech for a Computer Associates reseller. I am biased towards their products, but will be the first to admit that they sometimes do suck. Hard. Like turbine of an F-16 hard.
If you'd like to, I will give you a format in which to present your Request For Quote (RFQ) to the various vendors.. It will give you an unbiased measuring tool for all that FUD and MarketingSpeak.
unless they've recently changed the AFSC's you'll be a 3c031 when you leave tech school.
I spent four years babysitting 3c0's in base communication centers from New Mexico to Turkey then Arizona..
best advice.. 1> Don't try to get stationed near your home for your first assignment. It almost never happens, so don't waste a slot on your dream sheet.
2> Volunteer for overseas. You'll be amazed how much fun you can have in a foreign land even when you don't speak the language. I highly reccomend Turkey. Very friendly locals, great food, excellent liquor, and 18 yr old drinking age. You also come away with great stories
3> Don't be a dorm rat! Get out and see whats out there, even if you are in North Dakota.
4> DO NOT (as in NEVER) think you're in love in tech school.. It can be a very high stress environment and I saw a few good people go down in flames for people they thought they loved only to be trampled on like some bad high school romance. Remeber, you may feel more responsible, it doesn't mean the people around you are.
5> when they're handing out your assignments at the end of tech school, and you have the option to trade, get on the DSN line at the school house and call your possible shop. Ask what they do. If the answer is "Comm Center Operator" or they mention running the MDT, try to swap with a less bright classmate. I saw brilliant potential admins turn into mush brained zombies after 8 hours in front of that damned MDT console.
6> when you have to page a technician because your stuff isn't working, be nice, we were probably asleep when you paged.
Patrick J Sliney, Former Senior Airman, Secure Communications System Journeyman (2e351), 49th Comm Squadron, Incirlik Airbase, Turkey I've got lots more info for making life just a little better. email me at slineyp (at ) hotmail (dot) com BOHICA!
Drop me a line with any Questions folks, it's free from me.
I had a customer who had the same damn problem exactly. Tech support a nightmare, no one at legartoo returning call. ick.
So when I installed their SAN and they asked me to integrate notworker with the SAN, i almost fell off the console laughing.. I sold them CA's BrightStor Enterprise Backup.. (I'd just been to their "training" the week previous. It's pretty damn close to ArcServe, minus a few bugs (including a few new ones, though)
Overall, they're happy. managing Solaris, HP-UX, and win2k with agents for their four(!) database platforms and such all from one console backing up to a pair of spectralogic libraries..
I Told You So.
It's what I think when my clients come back to me after realizing NAS as a scalable solution doesn't exist. Rarely is there uprgrade ability in a NAS unless you buy the REAL expen$ive pieces like NetApp... Then you might as well just bend over and hand 'em your wallet.
The scalable answer is a SAN and no, they don't have to cost an arm or a leg.. We build 'em here with quality parts, the same stuff you get from Dell, without their pretty name all over it. I understand the questioner my not need a the huge capacity a SAN may give, but this is just another way to illustrate the point that NAS is a band-aid (at best) on a 5" gash when it comes to large capacity customers.
Oddly enough, my father just bought out a storage unit that contained a few (total 10, 2 NIB) touchscreen CRT's. They're Elo touch screens, older model, 15" Drop me a line and I can give you more details, maybe arrange some testing. He's pretty hot to get rid of his inventory. BTW, anoyone want a cisco 1601 w/ T1 card? (was also in that storage unit.) address: slineyp at hotmail dot com. I am a private party, not a retailer or spam-meister.
Slashdot Mirror
As long as we're being honest...
I was just laid off from Xiotech,(with about 100 others) one of the smaller SAN vendors..
But we call the Centerra a "data jail". It's like the roach motel..
Data checks in, but it don't check out. It can't scale beyond a 42U rack enclosure. It's a bunch of little servers striped together to form a big NAS with a metedata controller in the middle.
Just bashing the 800lb gorilla...
OTOH, If you're hiring, I'm willing to tell you your products could rule the world!
Regards.
Patrick
go to suprnova.org and search under linux apps, there should be a torrent there.
I second that..
...Win Friends ...
I just finished reading
And I AM a geek. I find it to be an excellent resource for dealing with people.
I mean all kind of people.
It was recommended to me by my mentor, and I can see him apply it almost daily.
Amazing resource.
The connection technology with the drive / spindle quality.
(P)ATA and SATA are connection technologies.
They have their individual benefits and drawbacks
(cost, reliability, speed)
The real factors to consider are the details of the drives themselves - vibration dampening, bearing and motor quality, MTBF.
It used to be rather simple to guess what quality of drive you were buying. If it was 146GB or less (73GB, 36GB), and rotational speed was 10K or 15K, it was either SCSI or FC, and an "enterprise" class drive, rated in Mean Time Between Failure.
Good drive, high quality, expect it to last several years, spinning 24 hours a day, sustaining high read and write activity during production and backup hours.
If the drive was larger (200GB+) and slower (7200 RPM), typically an ATA drive, maybe low end SCSI.
Then it was, at best, a workstation class drive, rated in "Contact Start Stops", meaning how many spin-ups and shutdowns the drive should survive. Not meant to run 24 hours a day, and run under heavy load except for short periods.
The lines are beginning to blur with 300 - 500 GB drives with FC drive attachment. Those drives are meant for archiving and reference data. Not production databases and such.
In my personal experience, the 3Ware products are worth the premium.
Pick your attachment technology as appropriate.
Best of Luck,
Patrick (slineyp at hotmail dot com)
Don't believe the marketing.
From what I read, the XSan software is first and foremost a distributed file system for shared volumes from the Xserve RAID.
If you look at the applications, it's about multiple servers or workstations with concurrent access to a single volume - distributed file locking.
Great stuff for the stated purpose, can't wait to get my hands on it!
Hiding the complexity of RAID is the domain of storage 'virtualization' solutions. The ones that let you mix and match raid types across any number of spindles you throw at it.
<Shameless_Plug>
My product, the XIOtech Magnitude does that. Take up to 126 spindles, create RAID 0, 1, 5, 10 volumes and give 'em to your servers. Boot off 'em, mirror 'em, copy 'em. Stick 'em in your ear!
</Shameless_Plug>
direct flames or questions to slineyp at hotmail dot com
When that slow 250GB ATA class drive is dead, and while its fellow drives are chugging their little hearts out (and probably maxing out that 3ware controller), how long will it take to rebuild your array?
Have you tested how long it takes? Probably better than 24 hours if your system is moderately loaded.
Guess what you have now? The marvelous opportunity for a CASCADING FAILURE!
That's right kids! Because you just had a drive fail, and all the other drives are doing double the work to rebuild from parity data, you have a higher chance of getting a second drive failure.
Consider that you bought all of the drives in that array at the same time. They've all been running the same amount of time. What if there was a minor manufacturing defect that caused that First drive to fail? How soon before it takes out the other 4?
A 'resume generating event' waiting to happen.
Best of luck.. and I agree with the comment upthread. SATA drives are for Workstations. Maybe for storing what we call 'reference data'.
Not much more.
There's a few choice terms in the industry- 'Economy Enterprise'
'Garbage RAID'
'Ghetto SAN'
Good luck
Damnit!
I remember the trailer you guys are talking about, but what was it for?
Does anyone have a copy?
I think it's your logic that's flawed skippy,
Equating an ORGANIZATION of racists to a inanimate object (video game) in terms of their ability to inflict actual damage IS unbelieveable.
I live in the south and I haven't seen an angry mob of GTA CD's hanging a "Negro" from a tree.
Moron
It takes a little setup, and a little cash, but I believe it makes for the best lab setup possible. I've designed several setups for customers like this:
- As many servers as you need (8, for example) Buy 8 identical servers - explained later.
- A Fibre Channel HBA for each host.
- A Fibre Channel disk array. Preferrably an old XIOtech Magnitude, available on eBay now for $14K - $20K, new from around $60K
- Connect up to 8 hosts directly to the Magnitude, assign a boot volume to the host from the Mag's management interface. Tell the host HBA to boot from the attached device
- Load the OS of your choice (most now support FC HBAs as boot devices). You can load multiple OS on each box, and swap logical hosts between physical boxes if need be.
- Lather, Rinse, Repeat.
email me with any additional questions- slineyp at hotmail dot com
I dealt with Atlantis once in Phoenix...
I told the recruiter about my OPEN SYSTEMS STORAGE background and the various HBAs, Switches, Tape, Disk, Virtualiztion, Management, etc.
She puts me on an interview with a guy looking for a wrench-turner who knows how to fix STK libraries..
It took about a minute for both of us to figure out she was wasting our time.
We spent the next 15 minutes bitching about ignorant recruiters.
Turns out HE interviewed with them a month ago.
phiff
You hate your Sixth.....
Lets take this offlist, I've got a question about the EVAs and the Auto RAID function
slineyp at hotmail dot com - thanks
but you are right about spreading your data and I/Os around.
The reason customers pay a large fortune for EMC / HDS systems is that they're designed to get ungodly performance (cache based RAID) and that near-mythical "five 9s" availability (redundant disks, cache, backplane, power, etc).
It seems to be a disagreement in design circles about using modular arrays like the EVAs as nearly a "Field Replacable Unit" versus using monolithic HA designs like the EMC Symm or HDS lightning.
Here Here!
I hope JNI folds soon.
I'm so glad QLogic is kicking their ass in the Sun market.
I'm also recommending QLogic -E(MC) certified cards to my EMC customers.
We're not held captive to that crap anymore!
The Emulex LPUTIL has to be the single least intuitive POS configuration utility I've ever seen in a Fibre Channel product (and that's saying a lot considering the Brocade mgmt interface)
disgruntled SAN d00d.
actually, I do have a clearance, though inactive.
Separated from Air Force in 2000 as a Crypto Maintenance tech.
Now I do SAN and DR stuff. Anyone hiring?
slineyp at hotmail dot com
Actually, you are right. I must have still been hazy after waking up.
Thanks for the correction.
They have half-depth 1U boxes. That's right, two servers in 1U, back to back.
Includes space between the two for cabling and cooling.
They specialize in delivering easy to manage (physically) racks of highly commoditized systems.
(I work with them in a reseller relationship)
Imagine a 71U rack(minus 1U for a switch), with 142 boxes, all dual proc. 248 procs in a rack!
Man, I wish they'd put the right link in there.
Hit the nail on the head, it's likely the encryption device, that sits between the radio input and the transmitter proper.
The reason they want it back is that is IS a CCI, and therefore by beuracratic law and common sense, must be recovered and accounted for.
As to the key not being intact, odds are the device was in use when crew perished. Likely they didn't zeroize (official term, no shit) the key. If the unit is intact, so is the key. Fortunately, the key storage space is "tamper proof" that would self destruct the storage area on any attempt to crack it open.
We just saw the same thing at the UAB (Birmingham, AL)
great, another frickin method of sending me ways to enlarge my schwanz or get a university diploma
If you have 3.5 TB AND the capital to get an autoloader, then by all means!!
Depending on your environment (you made no mention of what OS's, where the capacity is concentrated, etc.) you should be able to call one of the b/u software vendors directly (CA, Veritas, Legato, Bakbone, CommVault, etc) or call one of their channel partners and start sniffing for a deal. I work in this sector and man, everyone is slugging it out and dropping their prices to bottom just to win business.
Also consider some storage management software like SRM- It will let you know EXACTLY what you are backing up, who's hogging space (by user and group) and where your capacity is allocated (an mis-allocated)
The short version is this: With some SRM tools and a good automated backup system, you WILL reduce your workload. I know, I set this stuff up for a living.
Disclaimer: I am a tech for a Computer Associates reseller. I am biased towards their products, but will be the first to admit that they sometimes do suck. Hard. Like turbine of an F-16 hard.
If you'd like to, I will give you a format in which to present your Request For Quote (RFQ) to the various vendors.. It will give you an unbiased measuring tool for all that FUD and MarketingSpeak.
slineyp (at) hotmail (dot) com
unless they've recently changed the AFSC's you'll be a 3c031 when you leave tech school.
I spent four years babysitting 3c0's in base communication centers from New Mexico to Turkey then Arizona..
best advice..
1> Don't try to get stationed near your home for your first assignment. It almost never happens, so don't waste a slot on your dream sheet.
2> Volunteer for overseas. You'll be amazed how much fun you can have in a foreign land even when you don't speak the language. I highly reccomend Turkey. Very friendly locals, great food, excellent liquor, and 18 yr old drinking age. You also come away with great stories
3> Don't be a dorm rat! Get out and see whats out there, even if you are in North Dakota.
4> DO NOT (as in NEVER) think you're in love in tech school.. It can be a very high stress environment and I saw a few good people go down in flames for people they thought they loved only to be trampled on like some bad high school romance. Remeber, you may feel more responsible, it doesn't mean the people around you are.
5> when they're handing out your assignments at the end of tech school, and you have the option to trade, get on the DSN line at the school house and call your possible shop. Ask what they do. If the answer is "Comm Center Operator" or they mention running the MDT, try to swap with a less bright classmate. I saw brilliant potential admins turn into mush brained zombies after 8 hours in front of that damned MDT console.
6> when you have to page a technician because your stuff isn't working, be nice, we were probably asleep when you paged.
Patrick J Sliney, Former Senior Airman, Secure Communications System Journeyman (2e351), 49th Comm Squadron, Incirlik Airbase, Turkey
I've got lots more info for making life just a little better. email me at
slineyp (at ) hotmail (dot) com
BOHICA!
If you call them, they'll give you pricing based on quantity, as an employee I bought one (older) unit under $1,000 US
Disclaimer: I used to work with these guys.
That said, it's a solid product, and they could use the support
netzerver.com
is an embedded linux on a single board system, capable of SCSI or IDE.
I have one here in the home office. Speaks Apple, Netware, SMB, HTTP, etc. easy config, real reliable, real cheap.
Quick and (not so) dirty, it works.
Drop me a line with any Questions folks, it's free from me.
I had a customer who had the same damn problem exactly. Tech support a nightmare, no one at legartoo returning call. ick.
So when I installed their SAN and they asked me to integrate notworker with the SAN, i almost fell off the console laughing.. I sold them CA's BrightStor Enterprise Backup.. (I'd just been to their "training" the week previous. It's pretty damn close to ArcServe, minus a few bugs (including a few new ones, though)
Overall, they're happy. managing Solaris, HP-UX, and win2k with agents for their four(!) database platforms and such all from one console backing up to a pair of spectralogic libraries..
I Told You So.
It's what I think when my clients come back to me after realizing NAS as a scalable solution doesn't exist. Rarely is there uprgrade ability in a NAS unless you buy the REAL expen$ive pieces like NetApp... Then you might as well just bend over and hand 'em your wallet.
The scalable answer is a SAN and no, they don't have to cost an arm or a leg.. We build 'em here with quality parts, the same stuff you get from Dell, without their pretty name all over it. I understand the questioner my not need a the huge capacity a SAN may give, but this is just another way to illustrate the point that NAS is a band-aid (at best) on a 5" gash when it comes to large capacity customers.
Oddly enough, my father just bought out a storage unit that contained a few (total 10, 2 NIB) touchscreen CRT's. They're Elo touch screens, older model, 15" Drop me a line and I can give you more details, maybe arrange some testing. He's pretty hot to get rid of his inventory. BTW, anoyone want a cisco 1601 w/ T1 card? (was also in that storage unit.) address: slineyp at hotmail dot com. I am a private party, not a retailer or spam-meister.