SDDC in pocket – virtualization, cloud, networking, storage, devops, cloudops
LAB for 2022 is ready. Last year the price of the energy go so high, that I must to turn off some of my servers. So they are not running 24×7 🙁 Usually I run 2x vCenters, NSX-T, VSAN, Cloud Director, vRO, vRLI, Tanzu, K8S etc.. and a lots of linux VM’s all for testing and learning purpose. Licenses I have are from vExpert and VMUG Advantage.
2022-01-21
LAB for 2021 is ready. Last year new home DC building has been build.
Usually I run 2x vCenters, NSX-T, NSX-V, VSAN, Cloud Director, vRO, vRLI etc..
and a lots of linux VM’s all for testing and learning purpose. Licenses I have are from vExpert and VMUG Advantage.
2021-01-17
Still in progress… not all HW is in the list.. for now just the CORE HW is in the list… and new homeDC building is being build ..
Networking:
Compute:
DELL R710
DELL R710
DELL R710
NUC1 – Test & Dev
VSAN ROBO:
ESX2 (whitebox)
ESX5
Storage:
NearStore
QNAP 653Pro
Networking:
Compute:
GOLEM – ESXi 6.5.0 – Build 10719125
Networking:
Compute:
ESX
Storage:
Synology DS210j
Once you hit the “Test” button you can go watch some movies. It’s take a time depend on your HW. For me it was couple of hours.
Testing was done by using Vdbench VM with input argument “-fvdb-8vmdk-100ws-4k-70rdpct-100randompct-8threads”
Each ESXi host:
In this saved .zip file are also the graphs with more details. Same what you can see on web http://<YOUR_BENCHMARK2_IP>:8080/results/
Just few of them..
Per disk
Almost in each graph section you can click and get full graphs.
Also on a web and in the .ZIP file are result in text form…
Copyright (c) 2000, 2016, Oracle and/or its affiliates. All rights reserved. Vdbench distribution: vdbench50406 Wed July 20 15:49:52 MDT 2016 For documentation, see 'vdbench.pdf'. 16:52:41.075 Created output directory '/root/vdbench/output' 16:52:41.106 input argument scanned: '-fvdb-8vmdk-100ws-4k-70rdpct-100randompct-8threads' 16:52:41.170 16:52:41.171 Adjusted default JVM count for host=localhost from jvms=1 to jvms=8 because of iorate=max and a total of 8 sds. 16:52:41.171 16:52:41.311 Starting slave: /root/vdbench/vdbench SlaveJvm -m localhost -n localhost-10-180314-16.52.41.043 -l localhost-0 -p 5570 -d86 16:52:41.333 Starting slave: /root/vdbench/vdbench SlaveJvm -m localhost -n localhost-11-180314-16.52.41.043 -l localhost-1 -p 5570 -d86 16:52:41.363 Starting slave: /root/vdbench/vdbench SlaveJvm -m localhost -n localhost-12-180314-16.52.41.043 -l localhost-2 -p 5570 -d86 16:52:41.385 Starting slave: /root/vdbench/vdbench SlaveJvm -m localhost -n localhost-13-180314-16.52.41.043 -l localhost-3 -p 5570 -d86 16:52:41.408 Starting slave: /root/vdbench/vdbench SlaveJvm -m localhost -n localhost-14-180314-16.52.41.043 -l localhost-4 -p 5570 -d86 16:52:41.432 Starting slave: /root/vdbench/vdbench SlaveJvm -m localhost -n localhost-15-180314-16.52.41.043 -l localhost-5 -p 5570 -d86 16:52:41.470 Starting slave: /root/vdbench/vdbench SlaveJvm -m localhost -n localhost-16-180314-16.52.41.043 -l localhost-6 -p 5570 -d86 16:52:41.504 Starting slave: /root/vdbench/vdbench SlaveJvm -m localhost -n localhost-17-180314-16.52.41.043 -l localhost-7 -p 5570 -d86 16:52:42.427 All slaves are now connected 16:52:45.108 Starting RD=run1; I/O rate: Uncontrolled MAX; elapsed=3600 warmup=1800; For loops: None Mar 14, 2018 interval i/o MB/sec bytes read resp read write resp resp queue cpu% cpu% rate 1024**2 i/o pct time resp resp max stddev depth sys+u sys 16:52:46.275 1 8869.00 34.64 4096 70.24 3.055 2.808 3.639 312.221 12.932 27.3 22.5 4.2 16:52:47.047 2 27285.00 106.58 4096 69.45 2.133 1.826 2.830 29.948 3.173 58.3 20.5 3.5 16:52:48.077 3 29975.00 117.09 4096 70.13 2.118 1.781 2.912 46.267 3.361 63.5 13.6 2.4 16:52:49.065 4 30425.00 118.85 4096 69.48 2.089 1.716 2.938 30.170 3.152 63.6 6.4 1.1 16:52:50.068 5 31824.00 124.31 4096 70.48 2.003 1.651 2.842 59.798 3.233 63.7 3.2 0.7 16:52:51.060 6 31044.00 121.27 4096 69.51 2.048 1.700 2.840 52.844 3.213 63.5 13.5 1.8 16:52:52.066 7 33020.00 128.98 4096 70.06 1.933 1.629 2.643 26.811 2.974 63.9 5.8 1.0 16:52:53.063 8 32574.00 127.24 4096 69.99 1.954 1.629 2.713 28.718 3.070 63.7 3.7 0.7 16:52:54.057 9 32823.00 128.21 4096 69.90 1.947 1.609 2.732 23.752 3.194 63.9 3.7 0.7 16:52:55.056 10 33120.00 129.38 4096 69.56 1.926 1.599 2.673 24.818 3.125 63.8 2.6 0.7 16:52:56.051 11 32846.00 128.30 4096 69.28 1.928 1.639 2.580 24.851 2.875 63.4 3.8 1.0 16:52:57.059 12 32356.00 126.39 4096 69.44 1.976 1.672 2.669 30.840 3.177 63.8 2.6 0.3 16:52:58.080 13 34012.00 132.86 4096 69.92 1.905 1.575 2.670 28.106 3.177 64.8 2.0 0.7 16:52:59.057 14 33399.00 130.46 4096 70.01 1.877 1.570 2.594 38.115 3.134 62.6 3.6 1.6 . .
Related parts:
HCI Bench – vSAN performance tool – Part 1: Installation
HCI Bench – vSAN performance tool – Part 2: Configuration
HCI Bench – vSAN performance tool – Part 3: Tests and Results
Default configuration
Home LAB starts to grow. More and more test lab VM’s, nested ESXi hosts etc. So I was looking for another node. Dont want to spent a fortune so the price was the key. I had also other minimum requirements as 32GB of RAM and power consumption. After few weeks of searching I decide to buy a Intel NUC.
Components added to my #HOMELAB:
Disclaimer: The Intel NUC is not on VMware’s official Hardware Compatibility List (HCL)
Intel NUC is the best piece of HW for HOME mini-LAB. But once you start grow you need more network bandwith. And one pNIC is not enough. After few sec of googling I find great page DevTTY. Where is a lot of information about “supported” USB NIC for your Intel NUC. I choose TP-LINK UE300 because I had a couple of TP-LINK (3210, 5426) switches and they are doing great job.
So I ordered TP-LINK UE300 USB 3.0 network card. Once it was delivered to me, I connected it to the NUC. Then download the driver from DevTTY and upload it to one of my datastore. No maintenance, reboot is needed. The only thing you need to do is to change the “Host Image Profile Acceptance Level” to “Community supported”.
Then just log in to the ESXi console and install VIB file with command: