Re: Userspace scheduler?

To: lvs-users@xxxxxxxxxxxxxxxxxxxxxx
Subject: Re: Userspace scheduler?
From: Lars Marowsky-Bree <lmb@xxxxxxx>
Date: Wed, 5 Dec 2001 18:09:43 -0500
On 2001-12-06T01:04:42,
   Julian Anastasov <ja@xxxxxx> said:

>       Not sure but I see some bad things: even if we use netlink
> for such purposes may be we have to enqueue the packets and to dequeue
> them after the user space has some time to get the decision.

Yes, this is what I feared, because we do not want to block on this.

> The other model is as the current one: the user space defines policy for
> scheduling and loads it to the kernel (-s WLC, for example).

This is not a serious option I fear.

> What policy the user space can propose? Examples? 

In this case, it is a rather complex load balancing decision based on constant
feedback from the next tier fed through a neural network and then chosing the
best server.

(You may shot me now; this wasn't my idea, I am just pondering how to
integrate this with LVS)

And besides, the current code is in C++.

    Lars Marowsky-Brée <lmb@xxxxxxx>

Perfection is our goal, excellence will be tolerated. -- J. Yahl

Attachment: pgpfPQEYhcvZ0.pgp
Description: PGP signature

<Prev in Thread] Current Thread [Next in Thread>