|
Ggggah I need to stay out of the Spiceworks forums. If I see Scott Alan Miller yap on about how Local Disk is the best thing since sliced bread and SAN is poo poo one more time I'm just going to steal the axe he wandered out of the dark ages with and execute him. The guy constantly harps about the network being a ~major failure point~ for a SAN, which is true but I kinda wonder what wildly hosed up networks he's been setting up if his SAN takes a poo poo due to network problems more than once every two years or so.
|
# ? Aug 29, 2012 22:09 |
|
|
# ? May 21, 2024 19:12 |
|
Boogeyman posted:I can tell you exactly what that is. Every month we have to run a process that looks at about billion records individually, does some stuff, then inserts a new copy of each of those records in another table. Moving from local disk to the SAN greatly increased the length of that process, and they're convinced that it's due to the latency. In that case, everyone's right! They're right (it is the latency), you're right (batches will improve things) and I'm right (they're using SQL wrong.) Without knowing more details, here's a couple bits: -Don't try to switch everything into one atomic select/update. This will make things explode. Log files will balloon and you'll run into lock/io contention. Design your SQL so you can process them 1-10k rows at a time. -Avoid row looping constructs, use all set based operations -Try to run this process more often, daily, hourly, hell run it as a trigger on insert, who knows what you need -Redesign this whole process cause it sounds dumb. Possible technologies that you may want include table partitioning, materialized views, bulk insert -Get your logs somewhere faster, local storage or force it onto ssd, raid 50 generally sucks for logs Nukelear v.2 fucked around with this message at 22:21 on Aug 29, 2012 |
# ? Aug 29, 2012 22:17 |
|
Rhymenoserous posted:The guy constantly harps about the network being a ~major failure point~ for a SAN, which is true but I kinda wonder what wildly hosed up networks he's been setting up if his SAN takes a poo poo due to network problems more than once every two years or so.
|
# ? Aug 29, 2012 22:29 |
|
I don't know if this is the right place to put this question, but here goes. My company needs a means of storing files with military levels of security/encryption. The files will probably never get over 1GB in size. I've been looking at the Ironkey 4GB S200, has anyone had any dealings with this? The reason for this is for government contracts. When you develop something for a government entity, they need proof that the files were never accessed by people that shouldn't be looking at them. Is there anything else you guys would recommend for computer illiterate people to use on a daily basis?
|
# ? Aug 29, 2012 22:32 |
|
Misogynist posted:I, for one, like to ensure that my systems that nobody can interact with because the network is down stay up and running in the event of a complete service outage! I'd say that but I'd probably get banned over there. The entire community worships the guy as some kind of genius storage/VM expert but half of the poo poo he spouts is absolutely archaic or dumb. Every time a medium business pipes in asking for a SAN solution he tries to convince them to do the old "Roll your own" with a 1U HP chassis a DAS and a copy of FreeNAS. I personally think that's great... for a project SAN or a lab. But I'll be hosed if I ever put anything in production where the last line of "Yell at people till it works" is in my office.
|
# ? Aug 29, 2012 22:34 |
|
^^^ Why would we he propose that? It combines all the failure points of a SAN, with all the failure points of DAS in addition to the 'roll-your-own' problems. Crazy. Rhymenoserous posted:Ggggah I need to stay out of the Spiceworks forums. If I see Scott Alan Miller yap on about how Local Disk is the best thing since sliced bread and SAN is poo poo one more time I'm just going to steal the axe he wandered out of the dark ages with and execute him. This guy is insane. We need a SAN.txt thread that can be filled with everything this guy posts. Propose we change the thread title to 'Enterprise Storage Megathread: SAN is never what you want, only what you get stuck with.' Nukelear v.2 fucked around with this message at 22:38 on Aug 29, 2012 |
# ? Aug 29, 2012 22:35 |
|
Nukelear v.2 posted:This guy is insane. We need a SAN.txt thread that can be filled with everything this guy posts. Propose we change the thread title to 'Enterprise Storage Megathread: SAN is never what you want, only what you get stuck with.' I'm so glad I'm not the only one that noticed. I mean I've been working with various SAN solutions over the years and I know the pitfalls. He never seems to hit any of the ACTUAL pitfalls and always ends up talking about how local disk will make IO operations on the host VMDK 1% faster or some dumb bullshit metric that no one will ever care about. I expect at any minute to find out that he's basing all of his iSCSI knowledge from when he tried to setup a iSCSI vlan on his old 100mb network. Nukelear v.2 posted:^^^ Oops my bad, not Freenas: Openfiler He even created a lovely acronym to describe his lovely device: SAM-SD (Scott Allen Miller - Storage Device) http://community.spiceworks.com/topic/99354-what-is-a-sam-sd Because no one had thought of slapping a das on a lovely old server and installing openfiler before. It needed it's own name. Rhymenoserous fucked around with this message at 22:44 on Aug 29, 2012 |
# ? Aug 29, 2012 22:39 |
|
What an asshat. I actually follow Mr. Miller on Twitter, I think because he is from Buffalo, and frankly his idiocy doesn't seem to leak out there, probably because he would get called out on it. Remind me to never visit the Spiceworks forums. Time to unfollow.
|
# ? Aug 29, 2012 23:08 |
|
I knew I recognized that name. I recently read an incredibly verbose article by him on why RAID 5--even with a hot spare--is dumb in sufficiently large arrays. I mean, I agree, but the article sucked. He just restated the same point over and over and over again.
|
# ? Aug 29, 2012 23:11 |
|
Rhymenoserous posted:
I know, I'll install Nexenta on a Proliant and call it a SAM-SD.
|
# ? Aug 29, 2012 23:20 |
|
The ken Rockwell of storage!
|
# ? Aug 29, 2012 23:23 |
|
Boogeyman posted:I can tell you exactly what that is. Every month we have to run a process that looks at about billion records individually, does some stuff, then inserts a new copy of each of those records in another table. Moving from local disk to the SAN greatly increased the length of that process, and they're convinced that it's due to the latency. How are your tempdb's laid out? This type of looping query hits them very hard, and it's easy to forget about them when thinking in terms of optimizing db and log drives. tempdb's should have their own drive, and you should split them, so there is one tempdb per cpu core.
|
# ? Aug 30, 2012 00:14 |
|
People hang out on Spiceworks forums?
|
# ? Aug 30, 2012 14:50 |
|
University domainmaster hasn't set up LDAP and kerb SRV records so I can't register my vFiler into our domain
|
# ? Aug 30, 2012 14:55 |
|
Rhymenoserous posted:He even created a lovely acronym to describe his lovely device: Jesus christ
|
# ? Aug 30, 2012 16:18 |
|
evil_bunnY posted:University domainmaster hasn't set up LDAP and kerb SRV records so I can't register my vFiler into our domain You don't even have to set anything up. Do they have you pointing at a non-MS DNS server or something?
|
# ? Aug 30, 2012 18:25 |
|
Spamtron7000 posted:You don't even have to set anything up. Do they have you pointing at a non-MS DNS server or something?
|
# ? Aug 30, 2012 18:49 |
|
EoRaptor posted:How are your tempdb's laid out? This type of looping query hits them very hard, and it's easy to forget about them when thinking in terms of optimizing db and log drives. tempdb's should have their own drive, and you should split them, so there is one tempdb per cpu core. There's a C# application that calls a stored procedure, and the stored procedure operates on one record at a time. It doesn't seem to put a lot of stress on tempdb as far as I've been able to tell. I do have tempdb on local storage (10 15K SAS in RAID 10), and it has 8 data files. The server has dual Xeon X7560s (8 cores each), I only created 8 files since I've heard from a few different sources that more than 8 is overkill.
|
# ? Aug 30, 2012 19:48 |
|
Boogeyman posted:There's a C# application that calls a stored procedure, and the stored procedure operates on one record at a time. It doesn't seem to put a lot of stress on tempdb as far as I've been able to tell. I do have tempdb on local storage (10 15K SAS in RAID 10), and it has 8 data files. The server has dual Xeon X7560s (8 cores each), I only created 8 files since I've heard from a few different sources that more than 8 is overkill. Our DBA noted that the recommendation is 1 tempdb per core, but there is probably a law of diminishing returns there somewhere. If you've already checked, you are probably okay, it's just super easy to end up with a query like: SELECT Y FROM tbl.Y WHERE IN (SELECT Z FROM tbl.Z) And this type of query will beat the poo poo out of tempdb. If you only have one of them, the locking contention goes through the roof, and even though the number of accesses is low, the performance tanks while all the threads wait for a lock. Separate tempdb's solve this, and your local storage of 10 spindles in raid 10 probably keeps latency super low for all of them. This just caught us out recently, so I wanted to note it in case the same thing applied to you guys.
|
# ? Aug 30, 2012 20:31 |
|
My new backup SAN is up and in testing. It's a SAN if it's using iSCSI right? Not bad for $7500 with a Supermicro SC847 - 36 drive chassis.
|
# ? Aug 31, 2012 19:31 |
|
I think you mean your backup SAM-SD. Right terms people...
|
# ? Aug 31, 2012 19:45 |
|
Joke's on you if you're on BE though.
|
# ? Aug 31, 2012 20:10 |
|
evil_bunnY posted:Joke's on you if you're on BE though. Backup Exec gets a lot of poo poo, and most of it is deserved. However I've been backing up 3TB to tape for the past few years with it, without issue. I expect a few hurdles with BE 2012, especially going to disk based deduplication, but because of the full feature set and lower cost, its still the logical choice over something else like AppAssure or Veeam.
|
# ? Aug 31, 2012 20:34 |
|
Nope, sorry, you don't get to administer your SAN right now, you just get to stare at this because stupid loving Java interfaces.
|
# ? Aug 31, 2012 20:34 |
|
I have a dedicated VM with Java 6 Update 29 just for that. My desktop kept getting infected when I tried to use 6U29+firefox to browse the web, and it doesn't play well in Chrome. Still better than the Hitachi or HP interfaces I came from.
|
# ? Aug 31, 2012 21:02 |
|
KS posted:I have a dedicated VM with Java 6 Update 29 just for that. My desktop kept getting infected when I tried to use 6U29+firefox to browse the web, and it doesn't play well in Chrome. Which is funny because Chrome has been the most reliable browser when it comes to getting to my array. Firefox derps up too much.
|
# ? Aug 31, 2012 21:32 |
|
Rhymenoserous posted:I'm going to make this suggestion again: Roll back to prior to the update that broke everything, then go get a real backup solution. Since this is all on a das rolling back your drivers shouldn't affect anything on the DAS itself. The answer is, sadly, both. Data rention was part of it, but it was always meant to serve media to various sources on demand. The idea was to have a RAID to prevent a drive failure, not a controller failure. Anyway, I've moved the discussion over here, if you want to continue to help/scold me http://forums.somethingawful.com/showthread.php?threadid=2801557&pagenumber=154&perpage=40#post407077156
|
# ? Aug 31, 2012 23:59 |
|
blackmanjew posted:I don't know if this is the right place to put this question, but here goes. I have used IronKeys very extensivly for the past 4 years. They are the real deal: FIPS certified, hardware encryption, and good speed. You may want to check out their Enterprise models since it features a "remote destory" option and other neat features labeled "silver bullet"
|
# ? Sep 1, 2012 00:29 |
|
Heyyyy whoever installed our Compellent controllers racked them in the wrong order and swapped the front covers. edit: Manually flipping the boot order on one of my two SC040 controllers so I can do a BIOS update Mierdaan fucked around with this message at 18:35 on Sep 1, 2012 |
# ? Sep 1, 2012 17:08 |
|
I inherited a LeftHand P4000, and what the gently caress, is it not possible to just get more trays? Have to buy a new controller as well?
|
# ? Sep 1, 2012 19:16 |
|
P4k is a module-based system, like equallogic.
|
# ? Sep 1, 2012 20:33 |
|
At least it looks like the Nimble guys have decided to let you scale up or out depending on what you need. They just added the option to add trays to their existing units and to also upgrade the controllers if yours are becoming computer bound.
|
# ? Sep 2, 2012 00:00 |
|
Is there any reason why someone essentially starting from scratch and purchasing 2-3 new VM hosts and needing storage wouldn't look at something like the new release of vSphere essentials plus and combining it with the internal storage in the new ProLiants? This is for a business at the medium end of SMB (150 users).
|
# ? Sep 2, 2012 18:12 |
|
Caged posted:Is there any reason why someone essentially starting from scratch and purchasing 2-3 new VM hosts and needing storage wouldn't look at something like the new release of vSphere essentials plus and combining it with the internal storage in the new ProLiants? This is for a business at the medium end of SMB (150 users). Vulture Culture fucked around with this message at 18:18 on Sep 2, 2012 |
# ? Sep 2, 2012 18:15 |
|
Caged posted:Is there any reason why someone essentially starting from scratch and purchasing 2-3 new VM hosts and needing storage wouldn't look at something like the new release of vSphere essentials plus and combining it with the internal storage in the new ProLiants? This is for a business at the medium end of SMB (150 users).
|
# ? Sep 2, 2012 18:41 |
|
The main problem with VSA originally was the limitations. A lot of those are resolved in the next release: vCenter not being able to run on the same environment, one VSA per vCenter, no ability to expand, had to use pristine hosts. I had not heard of any reliability concerns. It is still overpriced solo, but I believe the bundle pricing is a lot more reasonable.
|
# ? Sep 2, 2012 19:40 |
|
Feklar posted:I have used IronKeys very extensivly for the past 4 years. They are the real deal: FIPS certified, hardware encryption, and good speed. I've had an 8GB personal for years too. They are good but lately the support has started to go downhill with them being 'in a partnership' with Imation since 2011. The onboard firefox has problems (like with addon settings being non-permanent) and the secure sockets managed to reliably crash a win 7 machine I was using at one clients place. But the core "secure files on a usb stick" is fine.
|
# ? Sep 3, 2012 13:56 |
|
Bravo, IBM, for magically correlating my SONAS callhome ticket with an email distribution list at Stony Brook University Medical Center and blasting logs of my system to half a dozen sysadmins at another organization. Icing on the cake: both our organizations deal with HIPAA patient data!
|
# ? Sep 3, 2012 18:40 |
|
Misogynist posted:Bravo, IBM, for magically correlating my SONAS callhome ticket with an email distribution list at Stony Brook University Medical Center and blasting logs of my system to half a dozen sysadmins at another organization. Icing on the cake: both our organizations deal with HIPAA patient data! ...what the gently caress?
|
# ? Sep 3, 2012 22:06 |
|
|
# ? May 21, 2024 19:12 |
|
Mierdaan posted:...what the gently caress?
|
# ? Sep 4, 2012 01:57 |