Due to my extreme busy schedule I haven’t blogged for a while. Besides studying for VCDX and preparing for VMworld I’m also involved in a couple of projects. One project is designing and implementing a vSphere 4 virtual infrastructure. The VI will host an Exchange 2010 environment. Due to the size of my client’s environment, 192 TB is used for hosting mailboxes. These datastores will be available thru RDM, which means creating 192 1-TB volumes and assigning them to every ESX host in the cluster.
I’ve tried to use the Centralized Management Console, but it’s tedious and error prone work. Mind numbing repetitive exercises makes me ask really dumb questions on twitter such as where did SCSI id 7 go? <homer>D’oh!</homer>. So to protect myself from further bashing and being ridiculed I started to search for the Lefthand CLI to be able to automate the creation and assignment of volumes on a Lefthand SAN.
After skipping both VMworld events in 2008 I’m attending the VMworld event in San Francisco. This is the first event after VMware released vSphere and I hope to see much in-depth information about the OS and it’s new features. Duncan Epping and Eric Sloof posted info about interesting sessions, so I started browsing the session catalog as well.
The following sessions seems to be very interesting; Continue reading
In my first post I had a question about the path data travels when sent to a “standby” virtual connect module.
To quote my own question :
“What will happen if the VMkernel decides to use that nic to send IO?
Is the Flexnic aware of the standby status of it “native” uplink? Will it send data to the uplink of the VC module it’s connected to or will it send data to the active uplink?
How is this done? Will it send the IO through the midplane or CX-4 cable to the VC module with the active uplink? And if this occurs what will be the added latency of this behavior?
HP describes the standby status as blocked, what does this mean? Will virtual connect discard IO send to the standby IO, will it not accept IO and how will it indicate this?”
Due to many performance studies about disk performance it is well known that disk alignment for both VMFS partitions and NTFS file systems improve IO performance such as reduced latency and increased throughput. Alignment of VMFS partitions are done when configuring storage via the VI client but aligning NTFS partitions in Windows system prior to Windows 2008 is a manual task.
Windows Server 2008 use a partition starting offset of 1,048,576 bytes (1,024 KB) for disk larger than 4GB. This provides a well enough alignment for most disks. According to the official documentation, windows 2008 uses a different partition starting offset for disks smaller than 4GB. Or as MS states in the document
Performance Tuning Guidelines for Windows Server 2008
“Note that Windows Server 2008 defaults to a smaller power-of-two offset for small drives.”
But which starting offset does W2K8 exactly use for smaller disks?
One of my clients bought a couple of HP blade c7000 enclosures recently. Including the new dual port Flex-10 mezzanine cards (nc532m) and Flex-10 Virtual Connect modules. Due to the fact that this technology is quite new, not much inside-info is found on the web. I’ve had lots of discussions with Ken Cline and Scott Lowe, which will publish an Flex-10 article by it’s own pretty soon. This write-up is a quick overview of lessons learned by me but even
more a call for answers.