Depending on how often you are sending messages back to the trackers and how often they are looking at the tube you may actually not have that many tubes in existence. 

Beanstalkd tubes only exist while there is someone watching them or there is a message in the tube.  So if nobody is looking at the tube and there are no messages for that tracker's tube - no tube will exist. 

Seems like a fairly reasonable use case now that you've explained it - would be interested to hear how it went after you build it. 


On Tue, Aug 19, 2014 at 8:18 PM, Keith Rarick <[email protected]> wrote:
On Tue, Aug 19, 2014 at 10:37 AM, Diego Bernardes
<[email protected]> wrote:
> 1) is beanstalk single thread?

Yes.

> 2) does beanstalkd have any hard limit on tube quantity?

No.

Note that when a client accesses a tube (such as a 'use' or 'watch'
command), beanstalkd does a linear scan through the list of all tubes
to find (or create) the named tube. If you have a very large number of
tubes, and clients that frequently switch between them, this might be
a problem, but it's worth measuring to be sure.

100,000 tubes doesn't sound like a lot.

> 3) whats the min size of a tube?

On my 64 bit Mac, the tube struct takes 552 bytes, plus there's an
8 byte pointer in the global tube list, so a minimum of 560 bytes for
an empty tube. Note that malloc often has overhead beyond this
(usually at most one or two machine words per object), and C
compilers have some leeway in how they lay out memory inside a
struct (aligning fields to address multiples of various constants), so
the true number for you can be different.

> 4) is beanstalk being used in any big company?

About two years ago, we had a thread about this. You can
see the public responses here:
https://groups.google.com/d/msg/beanstalk-talk/A1hIa9dx7AI/idvXpc3VL6cJ

Also, beanstalkd has grown in popularity since then.

Additionally, there have been bigger companies that have
told me they use and like beanstalkd but haven't chosen to
mention that fact publicly, for various reasons.

> dont want to migrate a large system to beanstalk and suffer from a project
> being discontinued or something like this.

These days beanstalkd maintenance is quiet, because it
mostly does what it needs to do.

I spend a little bit of time to merge important bug fixes, such
as the recent https://github.com/kr/beanstalkd/pull/221. I know
I'm not very responsive these days; if that's important to you
then you might consider sticking with another tool, but there are
no plans to discontinue beanstalkd or anything like that.

> i think beanstalk is awesome and does a amazing jog in what it propose.

Thanks!

--
You received this message because you are subscribed to the Google Groups "beanstalk-talk" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/beanstalk-talk.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "beanstalk-talk" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/beanstalk-talk.
For more options, visit https://groups.google.com/d/optout.

Reply via email to