JANET and cam.ac.uk cutting Usenet News feed from July
in [UK Mac]
|
Prev: Mail and trash
Next: Mac Pro problem
From: Jaimie Vandenbergh on 13 Apr 2010 06:48 On Tue, 13 Apr 2010 10:39:44 +0100, Jim <jim(a)magrathea.plus.com> wrote: >On 2010-04-10, Jim <jim(a)magrathea.plus.com> wrote: >> >> Having said that, I wonder what the actual figures are. I've got a >> recent email from news.individual.net so I might try asking them and see >> if I get a reply. > >And I did. > >Roughly - very roughly - they receive about 1.8-2.5GB a day. This includes >duplicates from multiple feeds, plus the occasional accidental binary. >They've also seen these numbers drop over the last few years. > >Weeding out the duplicates I _think_ they see something in the region of >half-a-Gig a day inbound. > >Somewhat less than I was expecting, to be honest. Even allowing for a year's >retention (no idea what their retention actually is) the space requirements >are well within off-the-shelf storage, although you'd also need redundency >on that storage. So worst case, on mirrored arrays, duplicated across a couple of physical locations, we're looking at... four 1Tb disks per year of retention. That's not going to be the reason to stop providing it, then! I can't imagine a text newsserver requiring more than five days a year of someone's time to keep running. The initial set up (nowadays) would be three days for kit+software, maybe five days for newsfeed contracts. But support queries could be of any volume. I can't even throw an estimate at that. The easy thing to do is make it an unsupported service, but from hazy memory I think JANET doesn't like doing that. Cheers - Jaimie -- "On two occasions I have been asked [by members of Parliament], 'Pray, Mr. Babbage, if you put into the machine wrong figures, will the right answers come out?' I am not able rightly to apprehend the kind of confusion of ideas that could provoke such a question."
From: Rowland McDonnell on 13 Apr 2010 06:49 Dr Geoff Hone <gnhone(a)globalnet.co.uk> wrote: > real-address-in-sig(a)flur.bltigibbet.invalid (Rowland McDonnell) wrote: [snip] > >The bandwidth use is trivial; such decisions are made by management, > >based on management rationalizations such as the claim you make. > No claim, Rowland! You have no idea of the system topology involved. No claim, Geoff! You have no idea what I know! You're just insulting me to get a rise out of me. [snip baseless nonsense] > >Therefore, any such claim is baseless opinion - > <snip> > The facts are mine, the baseless opinion is yours. And yet you have presented precisely no hard figures; sorry, I know I shouldn't argue with a troll, but I'm an annoyed by your insults. Rowland. -- Remove the animal for email address: rowland.mcdonnell(a)dog.physics.org Sorry - the spam got to me http://www.mag-uk.org http://www.bmf.co.uk UK biker? Join MAG and the BMF and stop the Eurocrats banning biking
From: David Sankey on 13 Apr 2010 10:50 In article <hun6s51mg862l5cesg649vkfan9pms6g9f(a)4ax.com>, Jaimie Vandenbergh <jaimie(a)sometimes.sessile.org> wrote: > On Sat, 10 Apr 2010 20:03:01 +0100, James Jolley <jrjolley(a)me.com> > wrote: > > >On 2010-04-10 19:14:37 +0100, Chris Ridd <chrisridd(a)mac.com> said: > > > >> On 2010-04-10 17:51:02 +0100, James Jolley said: > >> > >>> > >>> I'm curious to know what the average usage would be now on modern > >>> computers with Binary type servers. > >> > >> It would be a frightening amount. > > > >That's why i'm curious. All those movies you know. Would be interesting > >to see how accurate server's retention rates are as well. > > Last figure I saw, from Easynews.com about two years ago, was > 4.5Tb/day and rising. JANET FAQ says 4 GB per day, 80% binaries, but no indication when this was. See <http://www.ja.net/services/news/faq.html#10>. Kind regards, Dave
From: Jaimie Vandenbergh on 13 Apr 2010 10:59 On Tue, 13 Apr 2010 15:50:40 +0100, David Sankey <David.Sankey(a)stfc.ac.uk> wrote: >In article <hun6s51mg862l5cesg649vkfan9pms6g9f(a)4ax.com>, > Jaimie Vandenbergh <jaimie(a)sometimes.sessile.org> wrote: > >> On Sat, 10 Apr 2010 20:03:01 +0100, James Jolley <jrjolley(a)me.com> >> wrote: >> >> >On 2010-04-10 19:14:37 +0100, Chris Ridd <chrisridd(a)mac.com> said: >> > >> >> On 2010-04-10 17:51:02 +0100, James Jolley said: >> >> >> >>> >> >>> I'm curious to know what the average usage would be now on modern >> >>> computers with Binary type servers. >> >> >> >> It would be a frightening amount. >> > >> >That's why i'm curious. All those movies you know. Would be interesting >> >to see how accurate server's retention rates are as well. >> >> Last figure I saw, from Easynews.com about two years ago, was >> 4.5Tb/day and rising. > >JANET FAQ says 4 GB per day, 80% binaries, but no indication when this >was. > >See <http://www.ja.net/services/news/faq.html#10>. Woah. I think I remember that FAQ from when I was in University and on a JANET feed - back in 1995. Which may be when the number was accurate! Given that many hundreds of full-size DVD image rips get posted daily into the alt.binaries.multimedia heirarchy, there's no way 4gig daily would cover a full feed. But JANET haven't done binaries for years - I don't think they did them (or only a selected few groups) in 1995. It might easily cover a text feed plus stray binaries, though. Cheers - Jaimie -- My swerver room, my patch panels. By the time they figure out why none of the ports on their floor box work anymore I'll be done, dusted and down the pub with a pint of something brewed with yeast that was smarter than they are. -- Matt S Trout, asr
From: Dr Geoff Hone on 13 Apr 2010 13:14
On Tue, 13 Apr 2010 11:49:46 +0100, real-address-in-sig(a)flur.bltigibbet.invalid (Rowland McDonnell) wrote: >Dr Geoff Hone <gnhone(a)globalnet.co.uk> wrote: > >> real-address-in-sig(a)flur.bltigibbet.invalid (Rowland McDonnell) wrote: >[snip] > >> >The bandwidth use is trivial; such decisions are made by management, >> >based on management rationalizations such as the claim you make. >> No claim, Rowland! You have no idea of the system topology involved. > >No claim, Geoff! You have no idea what I know! You're just insulting >me to get a rise out of me. It is true to say that I have no idea what you know. But, unless you can name the three establishment to which I was referring, then I will have to conclude that my statement "You have no idea of the system topology involved" is accurate, and that your knowledge of this particular set of circumstances is minimal > >[snip baseless nonsense] How you can conclude that hard fact is "baseless nonsense", when I was there and you - almost certainly - were not, would challenge any logical mind. > >> >Therefore, any such claim is baseless opinion - >> <snip> >> The facts are mine, the baseless opinion is yours. > >And yet you have presented precisely no hard figures; sorry, I know I >shouldn't argue with a troll, but I'm an annoyed by your insults. Oh, so you need more than a statement that download bit-rates could drop into two figures. OK, how about this: a bit-rate of 54 bits/sec was frequently observed. Is that hard enough for you or do you want to know if I am using Decimal or Hex as a numbering system? > >Rowland. > > >-- >Remove the animal for email address: rowland.mcdonnell(a)dog.physics.org > Sorry - the spam got to me >http://www.mag-uk.org http://www.bmf.co.uk >UK biker? Join MAG and the BMF and stop the Eurocrats banning biking |