#778: "shares of happiness" is the wrong measure; "servers of happiness" is
better
--------------------------------+-------------------------------------------
Reporter: zooko | Owner:
Type: defect | Status: new
Priority: critical | Milestone: undecided
Component: code-peerselection | Version: 1.4.1
Keywords: reliability | Launchpad_bug:
--------------------------------+-------------------------------------------
Comment(by davidsarah):
> If {{{k_e / k}}} is not an integer, the minimal {{{m_e}}} is a little
more complicated. In that case, most of the servers must have {{{1 + k_e
// k}}} shares, but {{{k-1}}} of them only need {{{k_e // k shares}}}.
> So {{{m_e = (n-k-1)*(1 + k_e // k) + (k-1)*(1 + k_e // k)}}}.
Rearranging gives {{{m_e = n - k + 1 + n * (k_e // k)}}}.
The first expression for {{{m_e}}} is wrong, even though the rest of the
argument and the rearranged version is correct. What you mean is:
... So {{{m_e = (n-(k-1))*(1 + k_e // k) + (k-1)*(k_e // k)}}}.
Rearranging gives {{{m_e = n - k + 1 + n * (k_e // k)}}}.
--
Ticket URL: <http://allmydata.org/trac/tahoe/ticket/778#comment:27>
tahoe-lafs <http://allmydata.org>
secure decentralized file storage grid
_______________________________________________
tahoe-dev mailing list
[email protected]
http://allmydata.org/cgi-bin/mailman/listinfo/tahoe-dev
