Re: (cloudiaas2016support-81) CBTOOL Scalability Questions

[Michael R. Hines]

Redirecting to cbtool...@googlegroups.com:

Think I had a typo there: Should have been: cb> aiattach hadoop default
default none none none sut=hadoopmaster->20_x_hadoopslave # without the
'templist' part

Not to mention, that you would also have to perform your own
investigation about what the approriate "load_factor" should be on each
cluster size. Obviously, the more VMs you add, you would also need to
increase the amount of work the cluster has to perform.

So, for example, you might do:

cb> aiattach hadoop default default none none none
sut=hadoopmaster->40_x_hadoopslave,load_factor=80000000

That would not only increase the number of slaves, but also increase the
amount of work of data for the mapreduce, in this case.

/*
* Michael R. Hines
* Platform Engineer, DigitalOcean.
*/

On 07/18/2016 01:25 PM, Michael R. Hines wrote:
> aiattach hadoop default default none none none
> templist=sut=hadoopmaster->20_x_hadoopslave