Large-scale production datacenter memory error report 2015

Analysis and modeling of new trends from the field

This academic report presents an analysis of Facebook's fleet of servers over the course of 14 months. The article explores the failure rate of servers related to factors such as server age, number of cores, DIMM capacity, and memory density. The document proposes methods for reducing failure rates for a baseline server, as well as identifying potential impediments to the technique.
  • Language: English
  • Released: June 2015
Study included in Corporate Account only.
More than 7,500
Recommended studies and dossiers
Show more
Recent studies and dossiers
You may also be interested in...
Statista is a great source of knowledge, and pretty helpful to manage the daily work.

Christof Baron about Statista CEO, MindShare Germany

Any more questions?

Get in touch with us quickly and easily.
We are happy to help!