ESXi free offers no support and certain features that you'll want in that environment cost money. A lot of the advanced features just shut off after 60 days on the free version.
To make it clear I am not trying to argue with you. It probably is an extravagant cost but I just think that often people like to devalue us as IT we should seriously consider it long and hard before we do it to others.
Seidkona on
Mostly just huntin' monsters.
XBL:Phenyhelm - 3DS:Phenyhelm
The bare-metal hypervisor comes with zero high availability. You can't vmotion (migrate a machine while it's powered on) between hosts, so if a host dies, you're *DOWN*. With paid vsphere, you get automatic recovery, and the hosts automatically move machines back and fourth between themselves to balance resources and such. The features get more and more complex and impressive from there.
The bare-metal just lets you turn on a host and put some VMs on it, and that's it. It also used to have some fairly restrictive caps on cores and memory (like 4 and 32gig? and that's for the whole host, not individual VMs), but I don't know if they did away with that.
The other thing was Hyper-V which is Microsoft's thing. Some people here use it and are fine with it. For me, it's a hypervisor that runs on top of a windows server with all the good and bad that comes with it. I'm not saying it's a shitty product, but I am saying there's a reason VMWare is the top of the heap.
Cog on
+2
Options
RandomHajileNot actually a SnatcherThe New KremlinRegistered Userregular
Yeah, I mean, look, the networking setup/redundancy is worth having someone else set it up. Things like the vMotion should be on this subnet, and the SAN should be spread out across three different ports for bandwidth/redundancy. I have a really good understanding of everything but the networking in our pretty basic VMware/SAN setup.
And VMware is worth every penny. The next time someone asks for a server, and you deploy it in two minutes, you won't care how much it cost.
Yeah I'm not trying to devalue their work, and I feel like the hardware costs might be a little inflated (do we need that specific SAN, is there cheaper?), but 30k for worker seems obscene for setting up vmware.
You're telling me I could learn vmware and earn my yearly salary by working a few days for 2 clients, and I'm done?
not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
Yeah I'm not trying to devalue their work, and I feel like the hardware costs might be a little inflated (do we need that specific SAN, is there cheaper?), but 30k for worker seems obscene for setting up vmware.
You're telling me I could learn vmware and earn my yearly salary by working a few days for 2 clients, and I'm done?
This is why so many of us go into consulting because, yeah, I am saying that.
Specialized knowledge is big money.
edit: This is incidentally why I have decided to cert up.
Seidkona on
Mostly just huntin' monsters.
XBL:Phenyhelm - 3DS:Phenyhelm
Yeah I'm not trying to devalue their work, and I feel like the hardware costs might be a little inflated (do we need that specific SAN, is there cheaper?), but 30k for worker seems obscene for setting up vmware.
You're telling me I could learn vmware and earn my yearly salary by working a few days for 2 clients, and I'm done?
This is why so many of us go into consulting because, yeah, I am saying that.
Specialized knowledge is big money.
edit: This is incidentally why I have decided to cert up.
This is fucking nuts.
20K price difference between VMWare and the Microsoft Hyper-V.
Not sure which, if any I should do.
We definitely need to upgrade because we have 10 year old machines. I'd like to virtualize because the headache goes away for a lot of what I'm doing.
So, either way we're spending that $10k just to upgrade our infrastructure.
I think I can just not take the router option and save another 10k they tacked on the end.
The veam backup I should probably take so our VMs are backed up.
I don't know if I can convince my boss to spend 30k+ on this shit though.
not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
Part of our infrastructure issues is when the power goes our we're basically scrambling to make sure our servers aren't going to die.
The generator causes our UPS units to freak the fuck out because someone had the bright idea to put a 1 phase residential natural gas generator on a 3 phase commercial electrical system.
GJ.
Instead of spending the extra $5000 to get the commercial version they decided to save costs and now they're going to need to spend 3K to rewire this building which they've put off for 5 years. There's already visible arcing going on.
No idea what to even do.
not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
0
Options
AthenorBattle Hardened OptimistThe Skies of HiigaraRegistered Userregular
Start by demonstrating to your company how much money they'd lose in the event of an outage.
Also, HyperV is a lot better than it was in ye old days, but I don't know if it supports to high availability stuff that VMWare does. We set up a setup like that on an IBM SAN at the last job - fiber channel connects, 10GB/sec switch, the works. But while we had failover, we did not have high availability - it would need to restart a server on a new VM Host.
Actually, I'm going to back up a bit - the first question you need to ask is how much uptime does your company want to pay for?
Yea, sometimes things sound expensive, because they are. $60,000 over the course of 5ish years on HA if something goes down so you don't actually have any downtime really isn't a lot of money
I don't know what kind of bullshit you tell those remote severs, but they let you in every gd time.
The fucking ninja spymaster of CLUs.
Aioua on
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
Oh there was still plenty of typing, and let me be very clear: Pre mudding, I typed looking at my fingers around 30 wpm. Post mudding I typed high/drunk/asleep/watching tv around 130wpm. Still can't spell for shit, cause you don't tend to type 'bs tiamat' at work very much, but I can fuck words up fast.
tintin was just for making sure I never had to loot/sac corpses, sac junk loot, auto-join the arena, some auto-responses to a few combat triggers in PVP, walking directions, and myriad utility triggers and actions.
Oh, and I subbed and colored text so combat read differently than the default to highlight the important shit, and my prompt was different depending on if I was in combat or out.
My MUD of choice in high school was Ages of Despair.
I got about 10 of my friends into it at one point.
After a while, some of us were good enough typists, that we could type notes to each other on our desks and read them just by watching the other person's fingers.
0
Options
RandomHajileNot actually a SnatcherThe New KremlinRegistered Userregular
Yeah, with that kind of margin, you're basically past HA even, into the FT realm, where even the RAM and CPU are kept in sync between hosts. Most of our VMs could be down for days and nobody would care. I say invest the money. You will be happier in the long run. When it comes time to replace/expand you can copycat what they did and save money then.
Have a few towers I need to set up. Have OSPF to deploy on the remaining flat networks on my perimeter. Have a farm show coming up that I'm up to my goddamn ears in work and stress over. Have an untold number of IT projects to get started on, that are basically crumbling away into ash while they get zero attention.
Start by demonstrating to your company how much money they'd lose in the event of an outage.
Also, HyperV is a lot better than it was in ye old days, but I don't know if it supports to high availability stuff that VMWare does. We set up a setup like that on an IBM SAN at the last job - fiber channel connects, 10GB/sec switch, the works. But while we had failover, we did not have high availability - it would need to restart a server on a new VM Host.
Actually, I'm going to back up a bit - the first question you need to ask is how much uptime does your company want to pay for?
99.99% is Google's standard. and look how much they pay for that. 99.9 or 99 is a lot more doable.
For what's worth though, my company just finished moving all our VMs off the bit of HyperV we have and onto ESXi.
Something to consider with HA though is whether this is a feature that can be provided by the application - what are you serving? Because a files/databases basically can be replicated and sharded out so that a hardware failure is at worst a momentary connection interrupt. Applications which know how to fail over will basically save you a lot of time and hassle.
EDIT: Also if you're worried about cost, then you pretty much need to be looking at Openstack. Possibly you should be doing that anyway since vSphere is actually kind of bad.
The bare-metal hypervisor comes with zero high availability. You can't vmotion (migrate a machine while it's powered on) between hosts, so if a host dies, you're *DOWN*.
You can vMotion in Hyper-V.
I don't think Hyper-V can keep you running in the event of a host crash though, it would have to restart the virtual machine on another host - meaning some downtime. But you can do live migration with no downtime of virtual hosts either just the storage of the host or the machine and storage. I think you need windows server 2012 for this out of the box.
Le_Goat you have no idea how much I needed that today.
Protip: if you let your patching lay fallow for five years, don't go and try and patch hunders of machines in a 12 hour outage window. Things were okay. Now everything knows it's broken.
Aioua on
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
After 8 hours of picking and running we're down to only 20 alerts more than usual.
Started at like 300 more than usual.
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
0
Options
AthenorBattle Hardened OptimistThe Skies of HiigaraRegistered Userregular
Whoops. Forgot to run my tar.gz command in a screen session or on a remote, local desktop. Hopefully cloning this entire system doesn't take forever...
Your email should not be a storage device for every file you get.
I'm not shitting you, I've seen someone email themselves files specifically so they can keep them in outlook.
I also currently deal with someone who puts the vast majority of the important documents she needs into calendar appointments...and only there. Like, she'll attach a word document onto an appointment for Tuesday at 10am, so when her outlook reminder comes up she works on that document. If she's not done with it by the time the allotted time she gave herself, she attaches the document to a new appointment in the future. Only when she decides she's done with the document does it get stored on a filesystem or document library somewhere.
Your email should not be a storage device for every file you get.
I'm not shitting you, I've seen someone email themselves files specifically so they can keep them in outlook.
I also currently deal with someone who puts the vast majority of the important documents she needs into calendar appointments...and only there. Like, she'll attach a word document onto an appointment for Tuesday at 10am, so when her outlook reminder comes up she works on that document. If she's not done with it by the time the allotted time she gave herself, she attaches the document to a new appointment in the future. Only when she decides she's done with the document does it get stored on a filesystem or document library somewhere.
Wow that's both clever and stupid at the same time.
I'm not even mad.
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
Clever yes, but wouldn't it be even more clever to have, oh I don't know, a link to the document where it's stored on the file server? Hell, I'd almost want to delete the meeting just to see her freak out and then explain why she shouldn't do that anymore. Of course that would mean getting bitched out, but it might prove a point.
While I agree that being insensitive is an issue, so is being oversensitive.
Trying to get a warranty replacement from HP is the shittiest fuck these days
What the hell is wrong with your site HP!? I just want a drive. why do you keep switching back and forth between two environments. Why do half of the links not work.
Aaaurugh.
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
life's a game that you're bound to lose / like using a hammer to pound in screws
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
2 GB was the file size limit of FAT32, that's why that was what it was.
Most PCs use NTFS now, so the file size is much larger, once it's past 4ish GB, you start hitting a lot of thrashing of data because the hard drive has to search for a few seconds to find what you're looking for in the file. It would actually be better off directly on the file system now-a-days.
not a doctor, not a lawyer, examples I use may not be fully researched so don't take out of context plz, don't @ me
Posts
And by love I mean get stabby.
To make it clear I am not trying to argue with you. It probably is an extravagant cost but I just think that often people like to devalue us as IT we should seriously consider it long and hard before we do it to others.
XBL:Phenyhelm - 3DS:Phenyhelm
The bare-metal just lets you turn on a host and put some VMs on it, and that's it. It also used to have some fairly restrictive caps on cores and memory (like 4 and 32gig? and that's for the whole host, not individual VMs), but I don't know if they did away with that.
The other thing was Hyper-V which is Microsoft's thing. Some people here use it and are fine with it. For me, it's a hypervisor that runs on top of a windows server with all the good and bad that comes with it. I'm not saying it's a shitty product, but I am saying there's a reason VMWare is the top of the heap.
And VMware is worth every penny. The next time someone asks for a server, and you deploy it in two minutes, you won't care how much it cost.
This is a clickable link to my Steam Profile.
You really don't want a free product for the load out you are talking about.
XBL:Phenyhelm - 3DS:Phenyhelm
You're telling me I could learn vmware and earn my yearly salary by working a few days for 2 clients, and I'm done?
This is why so many of us go into consulting because, yeah, I am saying that.
Specialized knowledge is big money.
edit: This is incidentally why I have decided to cert up.
XBL:Phenyhelm - 3DS:Phenyhelm
This is fucking nuts.
20K price difference between VMWare and the Microsoft Hyper-V.
Not sure which, if any I should do.
We definitely need to upgrade because we have 10 year old machines. I'd like to virtualize because the headache goes away for a lot of what I'm doing.
So, either way we're spending that $10k just to upgrade our infrastructure.
I think I can just not take the router option and save another 10k they tacked on the end.
The veam backup I should probably take so our VMs are backed up.
I don't know if I can convince my boss to spend 30k+ on this shit though.
The generator causes our UPS units to freak the fuck out because someone had the bright idea to put a 1 phase residential natural gas generator on a 3 phase commercial electrical system.
GJ.
Instead of spending the extra $5000 to get the commercial version they decided to save costs and now they're going to need to spend 3K to rewire this building which they've put off for 5 years. There's already visible arcing going on.
No idea what to even do.
Also, HyperV is a lot better than it was in ye old days, but I don't know if it supports to high availability stuff that VMWare does. We set up a setup like that on an IBM SAN at the last job - fiber channel connects, 10GB/sec switch, the works. But while we had failover, we did not have high availability - it would need to restart a server on a new VM Host.
Actually, I'm going to back up a bit - the first question you need to ask is how much uptime does your company want to pay for?
http://royal.pingdom.com/wp-content/uploads/2015/04/pingdom_uptime_cheat_sheet.pdf
99.99% is Google's standard. and look how much they pay for that. 99.9 or 99 is a lot more doable.
99.9 is way too long, if we're done for a day, that would've easily paid for almost 1/8th of the cost of this thing.
Throw money at that dream.
XBL:Phenyhelm - 3DS:Phenyhelm
I don't know what kind of bullshit you tell those remote severs, but they let you in every gd time.
The fucking ninja spymaster of CLUs.
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
My MUD of choice in high school was Ages of Despair.
I got about 10 of my friends into it at one point.
After a while, some of us were good enough typists, that we could type notes to each other on our desks and read them just by watching the other person's fingers.
This is a clickable link to my Steam Profile.
Nope, fiddle-fucking with email today.
Hoo-fucking-ray.
For what's worth though, my company just finished moving all our VMs off the bit of HyperV we have and onto ESXi.
Something to consider with HA though is whether this is a feature that can be provided by the application - what are you serving? Because a files/databases basically can be replicated and sharded out so that a hardware failure is at worst a momentary connection interrupt. Applications which know how to fail over will basically save you a lot of time and hassle.
EDIT: Also if you're worried about cost, then you pretty much need to be looking at Openstack. Possibly you should be doing that anyway since vSphere is actually kind of bad.
XBL:Phenyhelm - 3DS:Phenyhelm
No wait, that's good.
As in Microsoft Failover Cluster?
Oh they do.
You can vMotion in Hyper-V.
I don't think Hyper-V can keep you running in the event of a host crash though, it would have to restart the virtual machine on another host - meaning some downtime. But you can do live migration with no downtime of virtual hosts either just the storage of the host or the machine and storage. I think you need windows server 2012 for this out of the box.
Protip: if you let your patching lay fallow for five years, don't go and try and patch hunders of machines in a 12 hour outage window. Things were okay. Now everything knows it's broken.
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
After 8 hours of picking and running we're down to only 20 alerts more than usual.
Started at like 300 more than usual.
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
I'm not shitting you, I've seen someone email themselves files specifically so they can keep them in outlook.
I also currently deal with someone who puts the vast majority of the important documents she needs into calendar appointments...and only there. Like, she'll attach a word document onto an appointment for Tuesday at 10am, so when her outlook reminder comes up she works on that document. If she's not done with it by the time the allotted time she gave herself, she attaches the document to a new appointment in the future. Only when she decides she's done with the document does it get stored on a filesystem or document library somewhere.
Wow that's both clever and stupid at the same time.
I'm not even mad.
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
What the hell is wrong with your site HP!? I just want a drive. why do you keep switching back and forth between two environments. Why do half of the links not work.
Aaaurugh.
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
Checkout hasn't worked for months for me, even in IE.
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
that's right we're on a fucked up cruise / God is dead but at least we have booze
bad things happen, no one knows why / the sun burns out and everyone dies
Most PCs use NTFS now, so the file size is much larger, once it's past 4ish GB, you start hitting a lot of thrashing of data because the hard drive has to search for a few seconds to find what you're looking for in the file. It would actually be better off directly on the file system now-a-days.