[Bioc-devel] Download stats for Feb/2011?

Luo Weijun luo_weijun at yahoo.com
Mon Feb 21 03:28:37 CET 2011


Hi Christian and Martin,
The same download pattern also occurred to my gage package. Actually it occurred to many non-core packages, i.e. those don't install through biocLite(). For those core packages or highly used packages already with thousands of downloads per month, this pattern is not seen. For instance, the biomaRt package you have mentioned, its February is similar to other normal months. Based on Martin's description, very likely some other computer cluster with a few hundred nodes is installing Bioconductor with all packages included. These exta few hundred downloads mean a lot more for the non-core packages, but little for core packages. 
A related question I am interested now, is there any way to know the total download stats, not only from FHCRC server, but also from the mirrors. Martin said mirror stats are not counted currently.
Weijun


--- On Sat, 2/19/11, cstrato <cstrato at aon.at> wrote:

> From: cstrato <cstrato at aon.at>
> Subject: Re: [Bioc-devel] Download stats for Feb/2011?
> To: "Martin Morgan" <mtmorgan at fhcrc.org>
> Cc: bioc-devel at stat.math.ethz.ch
> Date: Saturday, February 19, 2011, 12:56 PM
> Dear Martin,
> 
> Thank you for your reply, you are right with the distinct
> IPs downloads. 
> The high  downloads of different packages is still
> conspicuous (a 
> statistical outlier) but I must agree that it is not so
> strange as the 
> biomaRt outlier.
> 
> Best regards
> Christian
> 
> 
> On 2/19/11 5:18 PM, Martin Morgan wrote:
> > On 02/19/2011 07:44 AM, cstrato wrote:
> >> Dear all,
> >>
> >> Looking at the download stats of my package xps I
> was astonished to see
> >> that it was downloaded 692 times which is twice as
> high as usually, but
> >> by only 57 distinct IPs which is about twice as
> low as usually.
> >>
> >> Looking at the download stats for other packages I
> realized that the
> >> same strange pattern of download stats, very high
> overall downloads vs
> >> low downloads by distinct IPs is true for many
> other packages.
> >>
> >> Does someone have an explanation for this strange
> download pattern in
> >> February?
> >
> > Worth remembering that these stats are summaries of
> the server logs at
> > FHCRC; they exclude mirrors and can reflect whatever
> weirdness the
> > internet might concoct.
> >
> > February is only half done so I think that's the
> distinct IP answer; the
> > total downloads could be any of a number of reasons,
> for instance
> > biomaRt was very popular in September 2010, when what
> appears from their
> > IP address to be many nodes in a HPC cluster somewhere
> in the US midwest
> > decided that they needed to repeatedly install
> biomaRt.
> >
> > A more complete answer will take a little digging...
> >
> > Martin
> >
> >
> >>
> >> Best regards
> >> Christian
> >> _._._._._._._._._._._._._._._._._._
> >>
> C.h.r.i.s.t.i.a.n   S.t.r.a.t.o.w.a
> >> V.i.e.n.n.a       
>    A.u.s.t.r.i.a
> >> e.m.a.i.l:        cstrato at
> aon.at
> >> _._._._._._._._._._._._._._._._._._
> >>
> >> _______________________________________________
> >> Bioc-devel at r-project.org
> mailing list
> >> https://stat.ethz.ch/mailman/listinfo/bioc-devel
> >
> >
> 
> _______________________________________________
> Bioc-devel at r-project.org
> mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel
> 






More information about the Bioc-devel mailing list