Hi Bhaskar,
On Thu, Nov 14, 2013 at 10:10:44AM -0500, Bhaskar Maddala wrote:
> > Worse, when applying consistent hashing on that, only two servers
> > got the load for several seconds, then two other ones.
>
> This is interesting. I did not do a time series distribution in my testing
> and this
Thank you for your comments, Willy.
> 1) "hash algorithm" => I realized that this naming is confusing
> because it's used in conjunction with the balance algorithm.
> In practice, both the terms "hash algorithm" or "hash function"
> are used, with the latter being much more common. So I changed
Hi Bhaskar,
OK I'm finally done with this. Having reviewed the existing code
allowed me to change my mind on a few points.
1) "hash algorithm" => I realized that this naming is confusing
because it's used in conjunction with the balance algorithm.
In practice, both the terms "hash algorithm
Hi Bhaskar,
I noticed a mistake I'm going to fix. I was a bit bothered by the fact
that we had to exclude the hash types from the LB_ALGO while setting a
balance algorithm, and realized that the bits you reused for the hash
method are themselves covered by BE_LB_ALGO, which is not appropriate
sinc
On Wed, Nov 13, 2013 at 03:59:20PM +0100, Willy Tarreau wrote:
> On Wed, Nov 13, 2013 at 09:41:31AM -0500, Bhaskar Maddala wrote:
> > > 2/4 :
> > > - I really like your principle of keeping the default settings
> > compatible
> > > with current implementations. That said, I still think that s
On Wed, Nov 13, 2013 at 09:46:19AM -0500, Bhaskar Maddala wrote:
> Hello Willy,
>
>I wanted to ask you permission, on whether it was ok to blog about this
> change, once it was committed on Tumblr's engineering blog [1]. I can do
> so after the changes are in and have your permission on it.
On Wed, Nov 13, 2013 at 09:41:31AM -0500, Bhaskar Maddala wrote:
> > 2/4 :
> > - I really like your principle of keeping the default settings
> compatible
> > with current implementations. That said, I still think that since 1.4
> > does not have the full avalanche with consistent hashing
Hello Willy,
I wanted to ask you permission, on whether it was ok to blog about this
change, once it was committed on Tumblr's engineering blog [1]. I can do
so after the changes are in and have your permission on it.
Thanks
Bhaskar
[1] http://engineering.tumblr.com/
On Wed, Nov 13, 2013 a
Hello Willy,
Thanks for your comments,
> 1/4 :
> - OK in principle. I have rearranged the doc changes in order not to
> overrun the 80-column limit. I also found that naming the new flags
> BE_LB_HALG_* was clearer than BE_LB_HF_* since they're only used to
> define the algorithm. I have
Hi Bhaskar,
sorry for the long delay. Yesterday I found some time to come back to
your patches.
First, I want to thank you for all the well documented work you've done
on this subject, that's really great.
I'm having some comments, both on the implementation and the code itself.
1/4 :
- OK in
Hello,
I filled in the test result for the remaining cases. It is the same link
as before [1]. The tests were executed with 3 data sets as follows (a) 10K
requests (includes duplicate host headers) (b) All unique host headers in
1M requests (c) and 250K requests (includes duplicate hosts headers
11 matches
Mail list logo
