Green HPC: An analysis of the domain based on Top500

TL;DR

The objective of this work is to analyze Top500 and Green500 data from several perspectives in order to identify the dynamic of the domain regarding its environmental impact and derive a predictive model for the weight of HPC sector within the horizon 2030.

Abstract

The demand in computing power has never stopped growing over the years. Today, the performance of the most powerful systems exceeds the exascale and the number of petascale systems continues to grow. Unfortunately, this growth also goes hand in hand with ever-increasing energy costs, which in turn means a significant carbon footprint. In view of the environmental crisis, this paper intents to look at the often hidden issue of energy consumption of HPC systems. As it is not easy to access the data of the constructors, we then consider the Top500 as the tip of the iceberg to identify the trends of the whole domain.The objective of this work is to analyze Top500 and Green500 data from several perspectives in order to identify the dynamic of the domain regarding its environmental impact. The contributions are to take stock of the empirical laws governing the evolution of HPC computing systems both from the performance and energy perspectives, to analyze the most relevant data for developing the performance and energy efficiency of large-scale computing systems, to put these analyses into perspective with effects and impacts (lifespan of the HPC systems) and finally to derive a predictive model for the weight of HPC sector within the horizon 2030.

Figures (7)

• Figure 1: lifespan of HPC systems in the Top500 (in years) from 1993. The red horizontal line means that 80% of the systems stay less than 2 years.
• Figure 2: The evolution of the Top 1 [Purple lines] supercomputer's performance metrics ($R_{max}, R_{peak}$) compared to the average evolution of all the Top500 supercomputer [Green lines] by date, based on Top500 along with Moore’s law projection.
• Figure 3: The performance ratio average between the Linpack $R_{max}$ and the theoretical $R_{peak}$ over time for the first Top500 system.
• Figure 4: Maximum Efficiency of the Green500 supercomputers by list date along with Koomey's law projection starting at two different periods [2014 and 2019].
• Figure 5: Maximum Energy efficiency growth Green500 systems by date distinguished by architecture type (homogeneous vs heterogeneous)