skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Why LINPACK (and STREAM) Aren?t the Answer.


Abstract not provided.

Publication Date:
Research Org.:
Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA)
OSTI Identifier:
Report Number(s):
DOE Contract Number:
Resource Type:
Resource Relation:
Conference: Proposed for presentation at the Technical Deep Dive with Hewlett Packard Server Division held March 16-17, 2015 in Palo Alto, California, United States of America.
Country of Publication:
United States

Citation Formats

Hammond, Simon David. Why LINPACK (and STREAM) Aren?t the Answer.. United States: N. p., 2015. Web.
Hammond, Simon David. Why LINPACK (and STREAM) Aren?t the Answer.. United States.
Hammond, Simon David. 2015. "Why LINPACK (and STREAM) Aren?t the Answer.". United States. doi:.
title = {Why LINPACK (and STREAM) Aren?t the Answer.},
author = {Hammond, Simon David},
abstractNote = {Abstract not provided.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = 2015,
month = 3

Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • Bosonic q-oscillators commute with themselves and so their free distribution is Planckian. In a cavity, their emission and absorption rates may grow or shrink-and even diverge-but they nevertheless balance to yield the Planck distribution via Einstein`s equilibrium method, (a careless application of which might produce spurious q-dependent distribution functions). This drives home the point that the black-body energy distribution is not a handle for distinguishing q-excitations from plain oscillators. A maximum cavity size is suggested by the inverse critical frequency of such emission/absorption rates at a given temperature, or a maximum temperature at a given frequency. To remedy fragmentation ofmore » opinion on the subject, we provide some discussion, context, and references.« less
  • Advancements in high-performance networks (Quadrics, Infiniband or Myrinet) continue to improve the efficiency of modern clusters. However, the average application efficiency is as small fraction of the peak as the system’s efficiency. This paper describes techniques for optimizing application performance on Linux clusters using Remote Memory Access communication protocols. The effectiveness of these optimizations is presented in the context of an application kernel, dense matrix multiplication. The result was achieving over 10 teraflops on HP Linux cluster on which LINPACK performance is measured as 8.6 teraflops.
  • Typically, all costs associated with disposal of hazardous waste are eliminated when the material is sold for reuse/recycling. In the future, out-of-pocket disposal costs can be expected to increase, and the market value of many materials found in wastes should rise as finite cheap virgin material sources are depleted. The recognition that natural resources will become increasingly scarce (perhaps similar to oil) has prompted the major oil companies to acquire major non-ferrous metal companies. In order to determine whether a serious marketing effort for recycling is justifiable, an accurate characterization of the must be made. Innovation in developing new applicationsmore » is essential unless the material is one that has been traditionally recycled. In the coming years, as both the value of our waste and its dispsal increase, much greater emphasis must be placed on the principle of non-waste technology.« less