Today, I was going through videos at VMworld 2010.  After  which I made few points out, which is something interesting to be aware  of.

 

Changes made in vMotion


1. Overhead  associated with vMotion is now carried by (VMM) monitor in ASYNC way,  till 4.0 it was synchronous (getting pages from monitor to kernel and  again sending it to network module)

 

2. Now from vSphere 4.1 onwards we can perform,

4 concurrent vMotions over 1GBps network
8 concurrent vMotions over 10GBps network

 

3. Faster maintenance mode evacuations

 

A host running with 15 ODD virtual machines now takes nearly 80 – 90  Secs to evacuate virtual machines when entering maintenance mode, this  is what is said.

 

Changes in Resource Management

 

Major focus was in vSphere 4.1  memory management. Whole idea here is  to control memory overheads. And this is achieved by memory compression  techniques.
Memory compression now happens at VM level as well at host level.

Earlier compression was done with 4K swap pages and was loaded in  compression cache, now the same is done at 2K swap pages. ESX will not  allocate any physical memory specially for compression cache now, all  operations are carried at VM memory level.

 

When referring to break out  Sessions, just came through  “Storage DRS”. New term in VMware storage  arena. But really good stuff. Not sure, when this will be launched.

 

The aim is to automate the tedious task of managing space  and I/O  load  across a group of storage devices by intelligent initial  placement  and  by using storage vMotion of virtual disks between data  stores. This is   analogous to how server DRS uses vMotion for CPU and  memory load   balancing.

 

http://www.vmworld.com/docs/DOC-4668