IPv6 Subnet Masks - CompTIA Network+ N10-007 Prof. Messer
-
@Dashrender Comparing how many Host per network can exist in IPv4 agains IPv6 the GAP is huge.
-
@melvinsilva said in IPv6 Subnet Masks - CompTIA Network+ N10-007 Prof. Messer:
@Dashrender Comparing how many Host per network can exist in IPv4 agains IPv6 the GAP is huge.
I suppose, but is that really a real thing that one cares/considers?
I guess I don't see anyone trying to make old school class A networks in IPv4 (16.7 million hosts), let alone wanting something even larger. I'm sure someone must use something larger than a /22, but it's not something I've ever seen.
-
@Dashrender You are right. From a LAN point of View, IPv4 is just fine. But for ISP IPv4 is run out of game, due the needs of more IPs for custumers and networks, here is where IPv6 is needed.
-
@melvinsilva said in IPv6 Subnet Masks - CompTIA Network+ N10-007 Prof. Messer:
@Dashrender You are right. From a LAN point of View, IPv4 is just fine. But for ISP IPv4 is run out of game, due the needs of more IPs for custumers and networks, here is where IPv6 is needed.
Sure, they need more hosts - but do they actually need larger networks? i.e. it's less efficient for them to use /22 type networks? Maybe - I guess I'm asking the question.
-
@Dashrender said in IPv6 Subnet Masks - CompTIA Network+ N10-007 Prof. Messer:
I suppose, but is that really a real thing that one cares/considers?
Yes, always has been. That's why companies pay an arm and a leg for big IPv4 segments. Getting even a handful of addresses is hard for a business today.
-
@Dashrender said in IPv6 Subnet Masks - CompTIA Network+ N10-007 Prof. Messer:
I guess I don't see anyone trying to make old school class A networks in IPv4 (16.7 million hosts), let alone wanting something even larger. I'm sure someone must use something larger than a /22, but it's not something I've ever seen.
That's because no one can afford it. They aren't available. A class A hasn't been available to purchased since the 1990s. Technically Classes haven't even existed since then. But modern equivalents to a Class A haven't been even an option, at any price, in forever. So you don't see people trying to get them because they are so scarce that they are gone.
-
@Dashrender said in IPv6 Subnet Masks - CompTIA Network+ N10-007 Prof. Messer:
@melvinsilva said in IPv6 Subnet Masks - CompTIA Network+ N10-007 Prof. Messer:
@Dashrender You are right. From a LAN point of View, IPv4 is just fine. But for ISP IPv4 is run out of game, due the needs of more IPs for custumers and networks, here is where IPv6 is needed.
Sure, they need more hosts - but do they actually need larger networks? i.e. it's less efficient for them to use /22 type networks? Maybe - I guess I'm asking the question.
They often can't get a /22. When they do, they have to fight tooth and nail to get it.
-
@Dashrender I think you are thinking that because of NAT, people don't need big address pools. And to some degree that is correct, but NAT is the necessary fallback measure that we all use because we all need bigger pools than are available. Twenty years ago, only companies were impacted by this. Today, every household has to deal with NAT because we needed it literally at the "every house" level! That's how much IPv6 matters.
-
@scottalanmiller said in IPv6 Subnet Masks - CompTIA Network+ N10-007 Prof. Messer:
@Dashrender I think you are thinking that because of NAT, people don't need big address pools. And to some degree that is correct, but NAT is the necessary fallback measure that we all use because we all need bigger pools than are available. Twenty years ago, only companies were impacted by this. Today, every household has to deal with NAT because we needed it literally at the "every house" level! That's how much IPv6 matters.
Correct, NAT is the reason I didn't consider this a real issue.
But if the goal is to get back to 1:1 real IPs to every device, then sure, IPv6 allows for that easily. But I still don't see single networks of /8 being around anywhere - sure HP/IBM/some other huge company might have a /8 or even larger assigned to the company level, but the chances that they will leave it with a netmask of /8, i.e. the network, seems extremely unlikely. -
@Dashrender said in IPv6 Subnet Masks - CompTIA Network+ N10-007 Prof. Messer:
@scottalanmiller said in IPv6 Subnet Masks - CompTIA Network+ N10-007 Prof. Messer:
@Dashrender I think you are thinking that because of NAT, people don't need big address pools. And to some degree that is correct, but NAT is the necessary fallback measure that we all use because we all need bigger pools than are available. Twenty years ago, only companies were impacted by this. Today, every household has to deal with NAT because we needed it literally at the "every house" level! That's how much IPv6 matters.
Correct, NAT is the reason I didn't consider this a real issue.
But if the goal is to get back to 1:1 real IPs to every device, then sure, IPv6 allows for that easily. But I still don't see single networks of /8 being around anywhere - sure HP/IBM/some other huge company might have a /8 or even larger assigned to the company level, but the chances that they will leave it with a netmask of /8, i.e. the network, seems extremely unlikely.You are leaping from "we struggle to get 5 addresses currently" to "I don't see many places needing 16 million." The lack of need for 16m doesn't somehow preclude the need for more than one. And it's the more than one that IPv6 solves.
-
@scottalanmiller said in IPv6 Subnet Masks - CompTIA Network+ N10-007 Prof. Messer:
@Dashrender said in IPv6 Subnet Masks - CompTIA Network+ N10-007 Prof. Messer:
@scottalanmiller said in IPv6 Subnet Masks - CompTIA Network+ N10-007 Prof. Messer:
@Dashrender I think you are thinking that because of NAT, people don't need big address pools. And to some degree that is correct, but NAT is the necessary fallback measure that we all use because we all need bigger pools than are available. Twenty years ago, only companies were impacted by this. Today, every household has to deal with NAT because we needed it literally at the "every house" level! That's how much IPv6 matters.
Correct, NAT is the reason I didn't consider this a real issue.
But if the goal is to get back to 1:1 real IPs to every device, then sure, IPv6 allows for that easily. But I still don't see single networks of /8 being around anywhere - sure HP/IBM/some other huge company might have a /8 or even larger assigned to the company level, but the chances that they will leave it with a netmask of /8, i.e. the network, seems extremely unlikely.You are leaping from "we struggle to get 5 addresses currently" to "I don't see many places needing 16 million." The lack of need for 16m doesn't somehow preclude the need for more than one. And it's the more than one that IPv6 solves.
OK sure, I might have been leading that direction, but the original post this all stems back to talked about
@melvinsilva said in IPv6 Subnet Masks - CompTIA Network+ N10-007 Prof. Messer:
@steve This is amazing how the amount of host per network increase in IPv6.
Which is the 16 million hosts Plus bit. Now that's not what he meant, but it's how I read it.
It's absolutely true that IPv6 gives us billions and billions of small networks worth of IPs(each of those smalls being likely larger than the full IPv4) -
@Dashrender said in IPv6 Subnet Masks - CompTIA Network+ N10-007 Prof. Messer:
Which is the 16 million hosts Plus bit.
You keep mentioning Class A - which is the 1990's 16m hosts class.
-
@Dashrender said in IPv6 Subnet Masks - CompTIA Network+ N10-007 Prof. Messer:
It's absolutely true that IPv6 gives us billions and billions of small networks worth of IPs(each of those smalls being likely larger than the full IPv4)
Right, both are just a flexible pool. One is just way, way bigger.
