I found a great deal of value reading through this April 12 2016 publication, knowing that I'll be using this as a handy reference in the future. Note this distinguished list of just some of the prominent authors that contributed to these 153 pages of goodness:
- Chris Wahl
- William Lam
- Christian Mohn
- Anthony Spiteri
- Frank Denneman
Download the 153 page PDF here.
Here's an excerpt from the Foreward, by Frank Denneman of PernixData:
This edition (version three) is the best of both worlds. It contains
informative blog articles with in-depth design descriptions, as well
as short tweet-sized snippets of advice. With over 10,000 copies
of the previous versions distributed, it has big shoes to fill. But we
know it is up to the task, and I hope you enjoy it
We are excited to announce the newest edition of the vSphere Design Pocketbook is now available for free download.
This book is a “must read” for IT professionals, as it provides best practice advice (in Tweet and blog format) from the industry’s top experts on how to best architect a virtualized data center.
Strategic Topics Covered by the vSphere Design Pocketbook 3.0:
• Host, Storage and VM configuration
• Cluster and vCenter design
• Network and Security design
Over 10,000 copies of the vSphere Design Pocketbooks have been distributed to date! Version 3 is sizzling hot, so download your copy now.
Here's some topics that I found are likely the most relevant for home lab enthusiasts. I've included some teasers, but you'll really want to read the whole section of the PDF, to be sure to get the full details and context.
Patrick Schulz (Section 1-6)
Recently we got some storage problems with a couple of ESXi
hosts and one of the recommendations of the VMware support
was to update the async drivers. I think that there isn’t much
awareness about those drivers as it actually should be, that’s why I
wrote this short post.
Steve Wood (Section 1-10)
You may be aware that if you do use Autodeploy, a SD or USB
device to boot your hosts then the scratch partition runs on the
Ramdisk which is lost after a host reboot for obvious reasons as its
Chris Wahl (Section 5-8)
Just be careful that you don’t abuse the logical processors by
creating a virtual machine that has more vCPUs than there are
physical cores available. A
Ather Beg (Section 5-12)
Also when configuring it, avoid pointing the ESXi host towards the
Windows domain FQDN. Instead, add each time source’s FQDN
or IP address individually to the configuration.
Christian Mohn (Section 6-9)
In my opinion, unless there are specific reasons to not use the
appliance, it should be the default choice for all greenfield
Niels Hagoort (Section 4-25)
In order to ensure optimal performance from you host hardware.
Power management should always be set to ‘high performance’.
Other considerations with regards to CPU C-states and Turbo
boost are noted in VMware documentation:
Servers with Intel Nehalem class and newer (Intel Xeon 55xx and
newer) CPUs also offer two other power management options:
C-states and Intel Turbo Boost. Leaving C-states enabled can
increase memory latency and is therefore not recommended for
low-latency workloads. Even the enhanced C-state known as C1E
introduces longer latencies to wake up the CPUs from halt (idle)
states to full-power, so disabling C1E in the BIOS can further lower
latencies. Intel Turbo Boost, on the other hand, will step up the
internal frequency of the processor should the workload demand
more power, and should be left enabled for low-latency, high-performance
workloads. However, since Turbo Boost can over-clock
portions of the CPU, it should be left disabled if the applications
require stable, predictable performance and low latency with
Bas van Kaam (Section 3-2)
- 0 to 12 milliseconds – Looking good, the lower the number
the better off you are.
- 10 to 15 milliseconds – Still acceptable in most cases, user
might notice a small delay.
- 15 to 20 milliseconds – Step up and take action, most of your
users won’t be happy.
- 20 to 25 milliseconds – Get your pen and paper out and shut
it all down
Here are some more interesting tools for you to have a look
at, they will either calculate your current IOPS load or help you
predict the configuration and IOPS needed based on your needs
- IOMeter – measures IOPS for a certain workload
- ESXTOP – specific to ESX, provides certain disk states, totals,
reads and writes.
- WMAROW – web interface, used to calculate performance,
capacity, random IOPS.
- The Cloud Calculator – web interface, disk RAID and IOPS
- Process Monitor – general analyses of IOPS
- Login VSI – VDI workload generator, simulate user activity on
- Supermicro SuperServer SYS-5028D-TN4T mini-tower available in 1U version featured in 3 node PernixData $10K give-away