Summary
We report performance measurements made on the 2-CPU Cray X-MP at ECMWF, Reading. Vector (SIMD) performance on one CPU is interpreted by the two parameters (r∞, n1/2), and we find for dyadic operations using FORTRAN r∞ = 70 Mflop/s, n1/2 = 53 flop. All vector triadic operations produce r∞ = 107 Mflop/s, n1/2 = 45 flop; and a triadic operation with two vectors and one scalar gives r∞ = 148 Mflop/s and n1/2 = 60 flop. MIMD performance using both CPUs on one job is interpreted with the two parameters (r∞, s1/2),1/2where s1/2 is the amount of arithmetic that could have been done during the time taken to synchronize the two CPUs. We find, for dyadic operations using the TSKSTART and TSKWAIT synchronization primitives, that R∞ = 130 Mflop/s and s1/2 = 5700 flop. This means that a job must contain more than ∼6000 floating-point operations if it is to run more than 50% of the maximum performance when split between both CPUs by this method. Less expensive synchronization methods using LOCKS and EVENTS reduce s1/2 to 4000 flop and 2000 flop respectively. A simplified form of LOCK synchronization written in CAL code further reduces s1/2 to 220 flop. This is probably the minimum possible value for synchronization overhead on the Cray X-MP.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Calahan, D.A., 1977: Algorithmic and architectural issues related to vector processors, Proc. Intl. Symp. Large Eng. Syst., New York, Pergamon.
Calahan, D.A. and W.G. Ames, 1979: Vector Processors: models and applications, IEEE Trans. Circuits and Systems, CAS-26, 715–726.
Hockney, R.W., 1977: Supercomputer architecture, in Infotech State of the Art Conference: Future Systems, F. Sumner (ed.), Maidenhead, Infotech Int., 277–305.
Hockney, R.W. and C.R. Jesshope, 1981: Parallel Computers - architecture, programming and algorithms. Bristol, Adam Hilger, pp 423. (Distributed in North and South America by Heyden and Son, Philadelphia).
Hockney, R.W., 1983: Characterizing computers and optimizing the FACR(R) Poisson-solver on parallel unicomputers. IEEE Trans. Comput., C-32, 933–941.
Hockney, R.W., 1984: The ni-method of algorithm analysis, in PDE Software: Modules, Interfaces and Systems, B. Engquist and T. Smedsaas (eds.), Amsterdam, Elsevier Science Publ. ( North-Holland ), 429–444.
Hockney, R.W. and D.F. Snelling, 1984: Characterizing MIMD computers: e.g. the Denelcor HEP, in Parallel Computing 83, M. Feilmeir, G.R. Joubert and U. Schendel (eds.), Amsterdam, Elsevier Science Publ., North-Holland, 521–526.
Hockeny, R.W., 1985: Performance characterization of the HEP, in MIMD Computation: HEP Supercomputer and its applications, J.S. Kowalik (ed.), Cambridge, MIT Press.
Kuck, D.J., 1978: The Structure of Computers and Computation, vol. 1, New York, Wiley, pp 611.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1988 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Hockney, R.W. (1988). (r∞, n1/2 s1/2) Measurements on the 2-CPU Cray X-MP. In: Hoffmann, GR., Snelling, D.F. (eds) Multiprocessing in Meteorological Models. Topics in Atmospheric and Oceanic Sciences. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-83248-2_5
Download citation
DOI: https://doi.org/10.1007/978-3-642-83248-2_5
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-83250-5
Online ISBN: 978-3-642-83248-2
eBook Packages: Springer Book Archive