My Home Lab

Virtualization Lab – Current Configuration

(Current as of 3-23-2020)

My Home lab consists of various items for testing. I have an R610 running a management environment (VSA, Jump Box, AD, connection server, Security Server, NVIDIA License Servers, etc.). It has dual E5620 Intel quad core processors. It also has 24GB of RAM and 5-146GB SAS drives in a RAID 5 configuration. Some of the VMs are running on local storage, most are running on a Synology 1517+ NAS over iSCSI. It works as a system to support management.

This is connected to the world using a pair of Dell 6224 gigabit switches switches stacked, plus each has 10GB link modules as well. I also use an old D-Link router I had sitting around for some of my sim-isolated networking. It’s a pretty standard networking setup, nothing really special to tell you about.

Another system in my arsenal is a Cisco C240-M3. It’s running the Intel E5-2640 procs at 2.50GHz with 6 cores each. The system has 64GB of memory in it and I have it loaded with 2-74GB SAS drives in a RAID1 for my boot volume and 3-146GB SAS drives in a RAID 5 for my storage volume. It is also connected to my Synology and an EqualLogic PS5000 iSCSI array The nice thing about this sever is that it supports the NVIDIA K1 and K2 cards should I need them for testing. I picked this up on ebay for about $500  + drives (http://www.ebay.com/itm/Cisco-UCS-C240-M3S-v02-Server-2x-E5-2640-2-50-GHz-6-Core-64GB-Dual-PS-No-HDD/262840746028). I’m working on phasing out the C240 for some more modern systems

I also have Dell VRTX running in the environment. It’s running both M610 and M620 blades. I typically won’t power it on with the rest of the system running right now because it will overload the branch circuit that its on and I don’t have another circuit near by.

GPU Configuration

Again I try to keep it up to date on vGPU versions, and it depends on what I am testing

Thank you ebay

Most of the equipment in my lab is surplus material I find on ebay. It’s good enough to prove the concepts and expand my knowledge which is important to me even if it’s not sufficient enough to run a production workload on.  Though from time to time I do get to use the latest and greatest as part of my job.

NVIDIA Jetson Lab

(Current as of 3-23-2020)

At GTC 18 NVIDIA provided attendees with a discount for purchasing a Jetson TX2 development kit. I decided to pick one up and start working with it. The specs for it are as follows:

ML/DL Development Lab

(Current as of 10-27-17)

My development lab is a system, named Deep Thought,  is what I built to learn new programming techniques specifically around machine learning and deep learning. It’s designed to be powerful enough to run CUDA based workloads and get me started with training and inference. It’s nothing to fancy but it works.

The OS for deep thought is: Ubuntu 16

The software specs are as follows:

• NVIDIA CUDA Toolkit
• Tensor Flow

I built the rig below in October of 2016 for about $630.

Configuration:

• Mother Board – MSI Z170-A PRO LGA 1151 Intel Z170 SATA 6Gb/s USB 3.1 ATX Intel Motherboard
  • Intel Z170
  • Core i7 / i5 / i3 / Pentium / Celeron (LGA1151)
  • DDR4 3200*(*OC)/ 3000*/ 2800*/ 2600*/ 2400/ 2133
• RAM – G.SKILL Ripjaws V Series 16GB (2 x 8GB) 288-Pin DDR4 SDRAM DDR4 2400 (PC4 19200)
  • DDR4 2400 (PC4 19200)
  • Timing 15-15-15-35
  • CAS Latency 15
  • Voltage 1.2V
• CPU – Intel Core i3-6100T Skylake Dual-Core 3.2 GHz LGA 1151 35W BX80662I36100T Desktop Processor Intel HD Graphics 530
  • 14nm Skylake 35W
  • 3MB L3 Cache
  • Intel HD Graphics 530
• GPU – GIGABYTE GeForce GTX 750 Ti 4GB WINDFORCE 2X OC EDITION GV-N75TWF2OC-4GI
  • Core Clock: 1059 MHz / Boost Clock: 1137 MHz
  • Dual-link DVI-I / DVI-D / HDMI x 2
  • WINDFORCE 2X Cooling System
• Power Supply – 750W Thermaltake Toughpower TPD-0750M – SLI / CrossFire Ready 80 PLUS Gold Certification Active PFC Power Supply
  • ATX12V / EPS12V
  • Modular
  • 80 PLUS GOLD Certified
  • 100 – 240 V 47 – 63 Hz