texteditor
Premium Buffalo-based Hosting
Isn't he a father? He should try and look for a stable, real job instead of continuing to try this flipping thing if only because it isn't a reliable, responsible income source
Virtual6.net Ceases Operations for unforeseen reasons :)
- Thread starter drmike
- Start date
Hi VPSBoard!
As a representative of Virtual6 I feel it's only right I step in and clear this up straight away. Although Ashley's name is "above the door" he will be taking a back seat with the day to day running of Virtual6, that's what he has employed me for, he essentially provides the funding only at present.
I intend to improve the brand name as best I can and hopefully I will be able to recuperate some of the past client relationships, from many of Ashley's other VPS brands, though understandably this will not be possible in all cases.
As texteditor mentioned, Ashley does indeed have a family to take off, he also works some 50+ hour weeks as an Electrical Engineer with his past employer (Pre VMPort.com), obviously this leaves essentially no free time for him to participate in much at all.
Please forward any concerns to our support desk and I will attempt to talk them through as best I can. I'm not looking for an argument here, people are justifiably angry and rightfully so, I'm just trying to do my job.
Many Regards,
Jamie Stevens.
Virtual6 Management.
As a representative of Virtual6 I feel it's only right I step in and clear this up straight away. Although Ashley's name is "above the door" he will be taking a back seat with the day to day running of Virtual6, that's what he has employed me for, he essentially provides the funding only at present.
I intend to improve the brand name as best I can and hopefully I will be able to recuperate some of the past client relationships, from many of Ashley's other VPS brands, though understandably this will not be possible in all cases.
As texteditor mentioned, Ashley does indeed have a family to take off, he also works some 50+ hour weeks as an Electrical Engineer with his past employer (Pre VMPort.com), obviously this leaves essentially no free time for him to participate in much at all.
Please forward any concerns to our support desk and I will attempt to talk them through as best I can. I'm not looking for an argument here, people are justifiably angry and rightfully so, I'm just trying to do my job.
Many Regards,
Jamie Stevens.
Virtual6 Management.
Okay Ashley. Sounds good, but why use the name Jamie now? Seems too, well, stupid.
Shall I press the button?
Last edited by a moderator:
DomainBop
Dormant VPSB Pathogen
Hi Domainbop!
Yes I am the same Jamie that worked for VMPort. Hopefully there may still be people around that remember me, I was with VMPort from the start.
And yes, we utilize RAID-1. There's no need for anything larger, we host a maximum of 16x 1GB VPS per node and from our extensive testing, this set up works well, allows us to lower costs and have our customers spread across more physical machines.
Everything is dedicated, including CPU. We don't operate a fair share policy etc.
drmike
100% Tier-1 Gogent
Why the ZOMBIE mode for Virtual6? I was thinking someone accidentally posted WHT offer somehow...
"Enterprise VPS
Offering 100% dedicated resources including CPU power, these packages are intended for serious hosting that's mission critical or resource hungry. Satisfaction guaranteed and online within one business day."
[SIZE=13.63636302948px]Enterprise + serious + mission critical = RAID1 ????[/SIZE]
Hi drmike!
Yes, for the reasons I listed above. RAID1 with 16 customers IMHO is more stable than RAID10 with 64+ customers (As with most providers using E3xxxx servers), though I guess its an administrators preference.
If you have any other questions please ask away, honesty is definitely one of my top priorities for this brand. Nothing about our service is hidden, we want the customer to know what they're buying, I refuse to hide things behind a TOS.
Last edited by a moderator:
No. RAID1 is still RAID1, you´ll be in deep shit with "mission critical" customers when (not if) it happens. I hope you have backups.
Hi blergh!
Like I said, it's down to personal opinion, RAID1 still gives us a level of redundancy and good enough performance per VPS, it's better than a single disk. Yes, backups are taken daily and stored off-site, though customers should of course have backups off their own.
You're right though, maybe the wording should be changed, as I would consider mission critical hosting something with HA myself, which is not what we're offering here.
Last edited by a moderator:
drmike
100% Tier-1 Gogent
RAID1 is running Rambo style. I do it or worse on my own personal stuff, but on production, never. Never do I say Enterprise or similar cash justification terms with such a disk choice in the mix.
As far as container loading, depends on virtualization and actual use as to what is sane to pack on your servers. E3's lack core count for big workloads. Reason people use them???? Newer CPU acceptance in their minds... Unsure really. People would do better on other CPUs, even if older CPUs --- especially lean startups.
Aside from those gripes, congrats on sticking in here, replying, etc.
Hi drmike!
I completely agree with pretty much everything you have just said.. One of the focuses of our business plan was to offer dedicated CPU, with a policy that's straight forward and easy to understand for everyone.
To offer a good enough percentage on even our lowest plans, using older quad core CPU's was the obvious option, since E3's just don't offer the power we need. It's decisions like this and the fact that RAID1 and backups was cheaper than 4x disks + with RAID10 that have made it possible for us to offer such a competitive yet sensible price.
And don't congratulate me just yet, I have a lot of work to do
Last edited by a moderator:
To be honest, there isn't a big difference between RAID1 and RAID10 in terms of redundancy. RAID10 is just two RAID1s striped. Losing two drives in the same mirror will destroy your data in both instances (assuming you are using 2 drive mirrors).
RAID6 and RAID7 are the safest because they are the only ones that are guaranteed to survive 2 drive failures but they normally lack in performance, especially write IOPS. RAID7 is rare and I've only seen it in place in ZFS systems (RAIDZ3). RAIDZ3/RAID7 can survive 3 disk failures.
Last edited by a moderator:
drmike
100% Tier-1 Gogent
Correct me here if I am wrong, but even with RAID-10 on the atypical 4 drive setup, folks are running fairly stupid. Why? Only 2 sets of stripes. One fails other drives get punched by activity and increases potential for failure. If the other stripe breaks, you are done.
RAID-10 would seem to be sane say on 6 drives / 3 stripe setups and larger.
I think your understanding of RAID10 is partially incorrect. It is one stripe on top of two sets of mirrored drives. See the diagram on http://en.wikipedia.org/wiki/Nested_RAID_levels#RAID_10_.28RAID_1.2B0.29.
The I/O on each device in the mirrored sets isn't hugely different when one of them is failed.
The I/O on each device in the mirrored sets isn't hugely different when one of them is failed.
Last edited by a moderator:
I actually just posted this as a guide:
The number of stripes doesn't make much of a difference in this case. Only the number of drives in each mirror. Therefore a 6-drive RAID10 would be much more reliable ONLY if they ran each RAID1 with 3 disks, but that is unusual since it would mean they would only have 2/6 disks worth of actual storage available.
Last edited by a moderator:
Jamie, you are Ash's cousin right?
Regarding the whole raid thing, if 1 drive brakes you replace it asap anyway and assume the array will die imminently if not, the various levels after that are just about performance or reduced recovery effort, i.e. raid 6 vs raid 10, 6 is easier to recover if you die 10 is a bit more complex but performs faster etc etc.
Raid 1 unless SSD cached is just f'ing ridiculous in 2015 for a VPS node though.
Regarding the whole raid thing, if 1 drive brakes you replace it asap anyway and assume the array will die imminently if not, the various levels after that are just about performance or reduced recovery effort, i.e. raid 6 vs raid 10, 6 is easier to recover if you die 10 is a bit more complex but performs faster etc etc.
Raid 1 unless SSD cached is just f'ing ridiculous in 2015 for a VPS node though.
Last edited by a moderator:
I meant dual quad core CPU's of courseHi drmike!
I completely agree with pretty much everything you have just said.. One of the focuses of our business plan was to offer dedicated CPU, with a policy that's straight forward and easy to understand for everyone.
To offer a good enough percentage on even our lowest plans, using older quad core CPU's was the obvious option, since E3's just don't offer the power we need. It's decisions like this and the fact that RAID1 and backups was cheaper than 4x disks + with RAID10 that have made it possible for us to offer such a competitive yet sensible price.
And don't congratulate me just yet, I have a lot of work to do
Last edited by a moderator:
DomainBop
Dormant VPSB Pathogen
if Ash hadn't necroed the dead I wouldn't be necroing a year old thread.
Virtual6 is back... http://www.webhostingtalk.com/showthread.php?t=1450384
This necroing of this thread sh*t is getting to be an annual event //rolls eyes
Ash on LET today:
Last edited by a moderator:
This thread is fascinating and deserved resurrecting. A VPS provider who things RAID 1 offers "a level of redundancy". I suppose it does, if you can have a negative level of redundancy. A sysadmin offering to work for £6/hour? It was two years ago, but that is not all that long ago. I live in a low cost country and could not survive on that (not without working very long hours, anyway).
People on OVH seem to accept the "it was all OVH's fault" story. I know OVH does not have great customer service, but is it that bad?
People on OVH seem to accept the "it was all OVH's fault" story. I know OVH does not have great customer service, but is it that bad?