|
It may have been talked about in the many pages in this thread that I haven't witnessed, but anyone got any quick advice: A machine with two major datasets (used purely for storage): Now, would it be faster to have two 4 drive RAIDZ(5) arrays, or just put all the drives (8) in RAIDZ2(6). Obviously RAIDZ2(6) would give better fault tolerance, but it's all going to be backed up so that's not super crucial. Just in case anyone has had the same dilemma, or maybe this is indeed a stupid question with an obvious answer, which is fine too. HalloKitty fucked around with this message at 13:10 on Mar 6, 2012 |
# ¿ Mar 6, 2012 13:06 |
|
|
# ¿ Apr 28, 2024 08:31 |
|
FISHMANPET posted:Two vdevs (each RAIDZ set is a vdev) are faster than one vdev, even when it's the same pool. The way RAIDZ works, is that each vdev is as fast as its slowest member, because for every read it has to read off of every drive in the vdev. When you add two vdevs the data is striped across both (essential a RAID 0) so it can read twice as fast. Thanks!
|
# ¿ Mar 12, 2012 00:37 |
|
adorai posted:We installed our new Nimble HF40 this week and I would be surprised if it has more than 3mm of clearance left in the rack. The longest SAN I have ever seen. The controllers in it are just so goddamn huge. It has to be one of the absolute worst systems for disk space to physical size ratio I've ever seen.
|
# ¿ Mar 31, 2019 13:20 |
|
Wicaeed posted:God help me, our company keeps deciding to go with this all-flash storage vendor named Kaminario. Well, Infosight is relatively new to HPE, being a Nimble product
|
# ¿ Apr 10, 2019 17:45 |
|
It burns
|
# ¿ Apr 24, 2019 18:47 |
|
evil_bunnY posted:Hello friends one of our research groups wants a (probably all flash) ~100TB box to serve ~1GB files over 10GBE/SMB who should we be talking to? I'm the EU since that probably matters. It's probably going to come down to cost, right? It's hard to know exactly how much things cost without getting quotes. I've some experience with Nimble, and it's pretty positive.
|
# ¿ May 3, 2019 16:29 |
|
devmd01 posted:Just did controller hardware upgrades on both of our Pure installs, a+ would have them replace production hardware in the middle of the day again. Zero downtime, zero issues. I've done this with Nimble, same experience.
|
# ¿ Jun 15, 2019 07:18 |
|
skipdogg posted:Anyone worked with HPE Primera yet? Pros? Cons? We got one for our lab, trying to figure out the future of our storage systems. I have, we have two 2-node Primera A650s with a bunch of 7.68TB SSDs. At the end of the day there's not a lot of difference from 3Par, which we also have, and I've worked with (two 7200s, two 7400s, all 2-node). Hell, even the HPE support people still refer to Primera as 3Par, and SSMC reports the Primera as a 3Par. If you've ever worked on a 3Par before, you'll be right at home. As far as I can tell, there are a few internal upgrades, and the caching system has been re-worked. One of the nice things they've done is build the service processor right in to the controller, so you don't need to deploy a separate appliance or buy the physical service processor, and the updating process has been dramatically simplified. They also have locked options out in SSMC (I imagine you can probably coax Primera in to doing anything you want using the command line) to encourage best practices, so you can't choose RAID 5 any longer, to give one example. They've clearly taken that approach from Nimble, which is really incredibly restrictive compared to the flexibility of 3Par's InformOS. If I'm completely honest, I haven't been blown away by the performance (although we could possibly be experiencing some latency due to us running synchronous replication, but I'm not convinced), and some of the old limitations are showing - for example deduped+compressed volumes still have a size limit of 16TiB (although supposedly that's getting changed soon.. and don't start talking to me about VVols, in my opinion there are still too many issues surrounding them). I'm also not seeing dedupe and compression rates that amaze me, about 1.3x or so. I think they could do with stealing some of Nimble's code in that area (which we also run). Is there any specific information you're looking for? Have you worked with 3Par before, or with HPE? HalloKitty fucked around with this message at 09:05 on Jul 10, 2020 |
# ¿ Jul 9, 2020 19:01 |
|
I'm curious, what's happening with the Compellant gear, then? I'm guessing they'll have to pack up their poo poo and go home. Has anyone used and compared the newest Unity with the newest SC kit? I guess it's a pointless question now
|
# ¿ Jul 13, 2020 16:08 |
|
shame on an IGA posted:Good morning storage havers, we've got three weeks to scrape Imgur. It's Waffleimages all over again
|
# ¿ Apr 20, 2023 15:55 |
|
Pile Of Garbage posted:were just Brocade devices with IBM logos stuck to them. Seems to be very common, HPE does the same thing with StoreFabric. I guess when you don't want to be responsible for developing something as critical as Fibre channel switches, you just resell Brocade gear, especially as they're such an established player.
|
# ¿ Apr 20, 2023 18:48 |
|
Pile Of Garbage posted:Brocade got acquired by Broadcom yeah? I haven't worked with them in over a decade but I'm guessing they're now probably much worse. I think the products are fine, but the rituals you have to perfom to get access to the firmware updates...
|
# ¿ Apr 20, 2023 19:34 |
|
|
# ¿ Apr 28, 2024 08:31 |
|
Anyone have experience with on-prem object storage? I'm thinking Dell ECS, Scality Artesca/Ring, Cohesity Ideally a hardware appliance I can just deploy and scale by throwing more of them in a cluster. Interested in hearing quotes people have had, just out of curiosity. It'll be for backup.
|
# ¿ Dec 6, 2023 11:29 |