[arin-ppml] Tax incentive for renumbering-related labor costs?
cgrundemann at gmail.com
Tue Oct 21 20:56:26 EDT 2008
Speaking as someone who managed a small ISP network (around 2000
customers) through several re-numberings of various scale, I will take
For a single /24 (very rough estimates)
1) Design time; deciding what and how to do it. ~5hrs (@ $75 - $100hr)
2) Configuration time; adding secondary IPs, redundant routes, etc to
prepare for the migration, adding NAT, port forwarding, etc, and then
removing the old configs after the migration. Assuming that the whole
/24 is broken all the way to /30s, ~10hrs (@ $75 - $100hr)
3) Notification and cooperation; working with customers (internal or
external) to set expectations and work through issues that arise.
Again assuming 64 distinct connections, ~10hrs (@ $50 - $75hr)
4) Opportunity cost (maybe I am using the wrong term); for an ISP,
there will likely be at least one customer lost, not sure how/if this
would translate to a campus setting.
$1125 - $1500 in Engineering cost
$500 - $750 in Customer Service cost
$1000 in Opportunity cost (mostly making this number up)
= $2625 - $3250
So call it $3k per /24. Does this sound reasonable at all?
On Tue, Oct 21, 2008 at 5:43 PM, Tom Vest <tvest at pch.net> wrote:
> On Oct 21, 2008, at 7:33 PM, Owen DeLong wrote:
>> The problem with such an estimate is that it assumes renumbering
>> effort is proportional to network size.
> The problem with not fielding such an estimate is that no sane
> politician would consider sponsoring it.
> Come on folks, I thought that my use of the terms like "plausible"
> and "ballpark" would be enough to earn me a pass on assumptions of
> ignorance about the real-world diversity of production environments...
> No one wants to even venture a guess? No one else has seen a
> consultant study on renumbering costs that could be mined for a
> barebones "plausible" estimate?
>> A /24 that is, for example, not referenced in external configuration
>> data (not containing nameservers, not in ACLs, not being used to
>> terminate VPNs) is much easier to remember than even a /28
>> that contains a bunch of VPN terminators that have lots of tunnels
>> configured to third parties.
>> On Oct 21, 2008, at 3:23 PM, Tom Vest wrote:
>>> Anyone care to suggest plausible "ballpark" renumbering-related labor
>>> costs for one IPv4 /24?
>>> I vaguely recall hearing about a couple of private studies estimating
>>> such costs...
>>> Anyone think that this kind of tax incentive would be sufficient to
>>> motivate returns to ARIN?
>>> What (if any) legal changes would be required to permit U.S.-based
>>> IPv4 holders to claim U.S. tax benefits like this for address space
>>> returned to ARIN?
>>> Obviously lots of variables, esp. if other jurisdictions/RIRs are
>>> considered, but is this worth investigating?
>>> You are receiving this message because you are subscribed to
>>> the ARIN Public Policy Mailing List (ARIN-PPML at arin.net).
>>> Unsubscribe or manage your mailing list subscription at:
>>> Please contact info at arin.net if you experience any issues.
> You are receiving this message because you are subscribed to
> the ARIN Public Policy Mailing List (ARIN-PPML at arin.net).
> Unsubscribe or manage your mailing list subscription at:
> Please contact info at arin.net if you experience any issues.
More information about the ARIN-PPML