File:Top500 supercomputers performance development.svg
| Uploaded by | Morn |
|---|---|
| Upload date | 2022-03-24T13:28:42Z |
| MIME type | image/svg+xml |
| Dimensions | 1620 × 900 px |
| File size | 79.6 KB |
Summary
| Description |
English: Geometric growth of supercomputers performance, based on data from top500.org site. Y Axis shows performance in GFLOPS. Red line denotes fastest supercomputer in the world at the time. Orange line denotes supercomputer no. 500 on TOP500 list. Dark blue line denotes total combined performance of supercomputers on TOP500 list.
Deutsch: Exponentielles Wachstum der Supercomputerleistung, basierend auf den Daten der top500.org-Site. Die y-Achse zeigt die Leistung in GFLOPS. Die rote Linie stellt den schnellsten Supercomputer der Welt zum jeweiligen Zeitpunkt dar. Die Linie in orange stellt den Supercomputer auf Platz 500 der TOP500-Liste dar. Die dunkelblaue Linie stellt die kombinierte Gesamtleistung aller Supercomputer der TOP500-Liste dar. |
| Date | |
| Source | Own work |
| Author | Morn |
| Other versions |
|
Licensing
I, the copyright holder of this work, hereby publish it under the following license:
 
|
This file is made available under the Creative Commons CC0 1.0 Universal Public Domain Dedication. |
| The person who associated a work with this deed has dedicated the work to the public domain by waiving all of their rights to the work worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law. You can copy, modify, distribute and perform the work, even for commercial purposes, all without asking permission.
|
Python source code
#!/usr/bin/env python3
from pylab import * # uses Matplotlib
import matplotlib.dates as mdates
import datetime
# Date First Last Sum
# Data is from HTML source at http://top500.org/statistics/perfdevel/
li = """1993-05-01,59.7,0.42,1128.57
1993-10-01,124,0.47,1493.35
1994-05-01,143.4,0.84,2317.01
1994-10-01,170,1.16,2732.24
1995-05-01,170,1.96,3927
1995-10-01,170,2.49,4784.34
1996-05-01,220.4,3.31,5892.24
1996-10-01,368.2,4.62,7981.24
1997-05-01,1068,7.67,12844.19
1997-10-01,1338,9.51,16898.12
1998-05-01,1338,13.39,22625.14
1998-10-01,1338,17.12,29367.6
1999-05-01,2121,24.7,39062.57
1999-10-01,2379,33.09,50938.56
2000-05-01,2379,43.82,64230.11
2000-10-01,4938,55.3,88082.45
2001-05-01,7226,67.78,108276.78
2001-10-01,7226,94.3,134977.51
2002-05-01,35860,134.3,222263.77
2002-10-01,35860,195.8,291814.16
2003-05-01,35860,241.4,370049.26
2003-10-01,35860,378.6,526740.41
2004-05-01,35860,606.9,812313.21
2004-10-01,70720,840.02,1128834.52
2005-05-01,136800,1166,1694886.64
2005-10-01,280600,1645.7,2299341.68
2006-05-01,280600,2026,2789465.47
2006-10-01,280600,2736.9,3527867.61
2007-05-01,280600,4031,4950600.48
2007-10-01,478200,5937.33,6977593.51
2008-05-01,1026000,8996.78,12152691.08
2008-10-01,1105000,12593.5,17373520.05
2009-05-01,1105000,17110,22640789.57
2009-10-01,1759000,20070,28006450.09
2010-05-01,1759000,24670,32434684.56
2010-10-01,2566000,31124.36,43786881.73
2011-05-01,8162000,40187.29,58930025.84
2011-10-01,10510000,50941.4,74069633.68
2012-05-01,16324751,60824.4,123417786.71
2012-10-01,17590000,76411,162139386.75
2013-05-01,33862700,96619,223654338.13
2013-10-01,33862700,117831.30,250080467.17
2014-05-01,33862700,133700,273763781.09
2014-10-01,33862700,153381,308850512.09
2015-05-01,33862700.00,164559,359296049.36
2015-10-01,33862700.00,206304,417807043.60
2016-05-01,93014593.88,286100,567353038.81
2016-10-01,93014593.88,349333,672112377.49
2017-05-01,93014593.88,432200,748700301.81
2017-10-01,93014593.88,548672,845120504.67
2018-05-01,122300000.00,715551,1210914864.08
2018-10-01,143500000.00,874800,1414955581.78
2019-05-01,148600000,1021000,1559575379.78
2019-10-01,148600000,1142000,1646887142.78
2020-05-01,415530000,1228000,2206134393.77
2020-10-01,442010000,1316840,2428761851.77
2021-05-01,442010000,1511000,2786058799.77
2021-10-01,442010000,1649110,3036861783.77
"""
x, y1, y2, y3 = [], [], [], []
for line in li.splitlines():
e = line.split(',')
y, m, d = [int(w) for w in e[0].split('-')]
f, l, s = [float(q) for q in e[1:]]
x.append(datetime.date(y, m, d))
y1.append(f)
y2.append(l)
y3.append(s)
figure(figsize=(18,10))
semilogy(x, y1, label = "#1", lw = 2, color = "red", marker = "d")
semilogy(x, y2, label = "#500", lw = 2, color = "orange", marker = "p")
semilogy(x, y3, label = "Sum", lw = 2, color = "blue", marker = "s")
legend(loc = "upper left")
xlabel('Year')
ylabel('Performance (GFlop/s)')
grid('on')
xticks([datetime.date(y,1,1) for y in range(1990, 2050)])
yticks([10**x for x in range(-1, 20)])
axis((datetime.date(1993,1,1),datetime.date(2023,1,1), .1, 1e11))
ax = gca()
ax.xaxis.set_major_formatter(mdates.ConciseDateFormatter(ax.xaxis.get_major_locator()))
savefig("top500_supercomputers_performance_development.svg")
show()