mirror of
https://github.com/lancachenet/monolithic
synced 2024-11-22 03:53:06 +00:00
1d6889e715
Information was taken from lancachenet/generic
32 lines
976 B
Markdown
32 lines
976 B
Markdown
# Hardware examples and throughput
|
|
|
|
## Introduction
|
|
|
|
This project is heavily reliant on the hardware that you run it on. The most
|
|
common complaint is "The cache is slow" and in nearly every case this can be
|
|
attributed to the hardware that the cache is being run on.
|
|
|
|
|
|
The following hardware specifications are real-world examples of hardware that
|
|
has been used to run this project, so you can get an idea of the type of
|
|
performance that you can expect from varying hardware setups.
|
|
|
|
|
|
|
|
## High Capacity
|
|
|
|
24 CPU core, 140GB memory, 8 x Samsung 850 Pro SSD in RAID6
|
|
Served ~2,800 uniques, peaking at 10Gbps
|
|
|
|
|
|
## Medium capacity
|
|
|
|
24 CPU core, 180GB memory, 5 x 900GB 15K SAS drives, plus 180GB SSD (lvmcache)
|
|
Served ~250 uniques, peaking at 1.8Gbps
|
|
|
|
## Low capacity
|
|
|
|
CPU: Quad-core AMD Phenom X4 II 945 (old!)
|
|
MEM: 16GB DDR3
|
|
Disk: 2x2TB RAID0 + 512MB SSD LVM Cache.
|
|
Without the LVM Cache the system will push around 300-400mb/s, with the SSD Cache pushing around 800-900mb/s
|