Re: [Lit.] Buffer overruns
From: Morten Reistad (firstname_at_lastname.pr1v.n0)
Date: 02/15/05
- Next message: Daniel Vogelheim: "Re: OpenOffice encryption insecure? (repost)"
- Previous message: jstevh_at_msn.com: "Re: I was right, surrogate factoring proof"
- Maybe in reply to: Brian Inglis: "Re: [Lit.] Buffer overruns"
- Next in thread: lynn_at_garlic.com: "Re: [Lit.] Buffer overruns"
- Reply: lynn_at_garlic.com: "Re: [Lit.] Buffer overruns"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Date: Mon, 14 Feb 2005 23:00:03 GMT
In article <dKqdnXZJ9KsaL43fRVn-iA@rcn.net>, <jmfbahciv@aol.com> wrote:
>In article <AridnTgBIZQTqI3fRVn-1g@rogers.com>,
> rpl <plinnane3REMOVE@NOSPAMyahoo.com> wrote:
>>jmfbahciv@aol.com wrote:
>>goin' out on a limb here, but methinks the same as "overbooking" in a
>>hotel, ie: in a 1000 room hotel there might be 1200 rooms booked for any
>>future date on the assumption that a certain amount will be noshows.
>>That's part of the art of hotel-reservations: knowing by how much to
>>overbook so you get a full hotel... if you actually do have more guest
>>than rooms then you have to shuffle some off to another hotel (usually
>>the competition which makes it a little embarassing).
Or, just having 200 seats for breakfast at that same hotel, because
you know half your guests go directly to meetings, and the rest is
nicely spread out during the morning.
A failure in capacity will manifest itself as a little queing, then
a little more queing; and then you had better rent the resturant next
door. But if the resturant is 1/3 full all morning you know you don't have
a problem.
>That's one. Think about file system designs that allow holey
>files. Or batch jobs that are going to be using 100 magtapes
>on a system that has two drives.
>
>There's all kinds of overcommits. Most of the time it all
>works until, e.g., both magtape drives barf all over the
>floor. Now you have a batch job bitching at the operator
>to mount tape FOO. He can't. So the operator has to deal
>with nattering TTY output. Most of the time under most
>OSes, there's an OPR command that stops, reschedules, and
>delays batch jobs that need magtapes but continue to
>start batch jobs that don't need magtapes.
Overcommits are everywhere. Last time I flew they had auctions
for whoever was willing to step off the plane. If they force you
off, they pay € 300 plus all expenses; but they can give anyone
an offer for € 100, and see if there are any takers.
-- mrr
- Next message: Daniel Vogelheim: "Re: OpenOffice encryption insecure? (repost)"
- Previous message: jstevh_at_msn.com: "Re: I was right, surrogate factoring proof"
- Maybe in reply to: Brian Inglis: "Re: [Lit.] Buffer overruns"
- Next in thread: lynn_at_garlic.com: "Re: [Lit.] Buffer overruns"
- Reply: lynn_at_garlic.com: "Re: [Lit.] Buffer overruns"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Relevant Pages
|