Skip to Content.
Sympa Menu

per-entity - Re: [Per-Entity] distribution of aggregate metadata

Subject: Per-Entity Metadata Working Group

List archive

Re: [Per-Entity] distribution of aggregate metadata


Chronological Thread 
  • From: IJ Kim <>
  • To: "Cantor, Scott" <>
  • Cc: Nick Roy <>, Chris Phillips <>, Per-Entity Metadata Working Group <>
  • Subject: Re: [Per-Entity] distribution of aggregate metadata
  • Date: Fri, 12 Aug 2016 12:02:26 -0400
  • Authentication-results: spf=none (sender IP is ) ;
  • Ironport-phdr: 9a23:RpY7hhBrSZh7oMJGv8OUUyQJP3N1i/DPJgcQr6AfoPdwSP3zr8bcNUDSrc9gkEXOFd2Crakb26yL6Ou5BCQp2tWojjMrSNR0TRgLiMEbzUQLIfWuLgnFFsPsdDEwB89YVVVorDmROElRH9viNRWJ+iXhpRZbIBj0NBJ0K+LpAcaSyp3vj6Hhs6HUNk9jjTyhZqk2ZC69qhnN/IFCioJkNqErjEHhpWBVPela2DU7C0iUmkPdxI+T/ZsrpyVSk/Mn68NaV6jmJeI1QaEOX2duCHw8+MC+7UqLdgCI/HZJFzxOyhc=
  • Spamdiagnosticmetadata: NSPM
  • Spamdiagnosticoutput: 1:99

On Thu, Aug 11, 2016 at 06:03:28PM +0000, Cantor, Scott wrote:
>
> I don't think I would be that concerned about basic CDN reliability in
> practice, *if* we're caching on disk. Which we're not at present. So
> that's really my take-away, that nobody by and large really believes
> networks and CDNs can meet anything above 3-4 9s. and semi-regular but
> unpredictable outages are inevitable.
>

It occurs to me that running multiple mdq metadata providers wouldn't be
too hard while I'm not sure if clients can chain them or can consume a
second mdq provider with something like "backingMDQUrl". Achieving five
9s among these providers would be much feasible.

--IJ




Archive powered by MHonArc 2.6.19.

Top of Page