New Home Server: Performance and Power Consumption

The following table shows the new and the old server components that affect the performance.

Part New 2015 Server Old 2009 Server
Motherboard Gigabyte GA-Z170X-UD3 LGA1151 Asus P5Q-EM LGA775
CPU Intel Skylake i7-6700 3.4GHz Intel Core 2 Duo E8400 3.0GHz
Memory 16GB 2133MHz DDR4 8GB 667MHz DDR2
System hard drive Samsung 850 Pro 512GB SSD WD Caviar Green 2TB SATA II 32MB

For the benchmarks my aim was to get a feeling how much faster the new server is. I ran the different benchmarks few times and I took the median value. I didn’t stop windows services, background tasks or anything like that so your mileage may vary.

Passmark

Fast, easy to use, PC speed testing and benchmarking. PassMark allows you to objectively benchmark a PC using a variety of different speed tests and compare the results to other computers. The results are expressed as PassMark Rating.

passmark.png

7-Zip

The 7-Zip benchmark shows a rating in MIPS (million instructions per second). Compression speed strongly depends from memory (RAM) latency, Data Cache size/speed and TLB. Decompression speed strongly depends on CPU integer operations.

7zip.png

3D Particle Movement

3DPM is a self-penned benchmark, taking basic 3D movement algorithms used in Brownian Motion simulations and testing them for speed. High floating point performance, MHz and IPC wins in the single thread version, whereas the multithread version has to handle the threads and loves more cores.

3dpm.png

Cinebench R15

Cinebench is a benchmark based around Cinema 4D.

cinebench.png

Rendering - PovRay

The PovRay, is a freeware package for ray tracing. It contains benchmark for stressing all processing threads on a platform.

povray.png

Power Consumption

For measuring power consumption I ran Handbrake which is capable of taking advantage of all the CPU cores. I used power consumption meter that was attached between the server’s PSU and wall plug to measure the power usage.

power_consumption.png

When idling the power consumption stays below 20 watts (usually 15-18W) which is amazing from a full blown server. It seems with these new CPUs you get best of the both worlds: High performance when needed and low power usage during idle time.

Conclusion

If you would upgrade the CPU every 1-2 years the performance benefits would be minimal. Something like 8-12%. But when upgrading six year old PC the difference shows. Depending on the benchmark the new server is 4-6 times faster than the old one. All that with less than 20W power usage during idle time.