Re: Ceph Performance very bad even in Memory?!

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



> 
> 
> 	SDS is not just about performance. You want something reliable for
> the next 10(?) years, the more data you have the more this is going to be an
> issue. For me it is important that organisations like CERN and NASA are using
> it. If you look at this incident with the 'bug of the year' then you see the
> importance of such organisations/large user base being involved.
> 
> 
> Sadly, that does not mean too much for me. From my personal experience I
> saw plenty of companies lacking behind

That is just demographics/statistics depending on where you are situated on the intelligence chart, you probably have to cope with the fact that quite a lot of people are below you and these people have jobs. So it is inevitable that you encounter such things. 

> by using state of the art technology
> and accepting huge software bottlenecks just by using excuses over and over
> again instead of deeply optimizing software/hardware components.

Yes some choose this, maybe best example is LTS versions.

> Same on installing security updates, using proper ci, ... Another example is
> that a lot of companies are still using openvpn (~70k+ lines of code,very bad
> performance with huge latency) vs wireguard (~4k lines of code, blazing
> fast).

I would not be able to conclude anything from just the amount of lines of code, other than one has more lines of code then the other.

Do you know by any chance if wireguard is able to run unprivileged and supports tcp? (I struggle with getting openvpn running in a unprivileged container and I need tcp for pass through in haproxy)
 
> 	Make sure to post results here. I am curious if ceph is really that far
> behind in performance compared to other solutions.
> 
> 

No idea, maybe Vitali can post graphics of results on his wiki. But link definitly here, I want to get the ceph devs also excited.
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx



[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux