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

Title: Experience on New Architectures with the Higrad Code

Authors:
 [1]; ORCiD logo [1]; ORCiD logo [1]
  1. Los Alamos National Laboratory
Publication Date:
Research Org.:
Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
Sponsoring Org.:
Defense Threat Reduction Agency (DTRA)
OSTI Identifier:
1345120
Report Number(s):
LA-UR-17-20718
DOE Contract Number:
AC52-06NA25396
Resource Type:
Conference
Resource Relation:
Conference: Programming Models and Co-Design ; 2017-02-06 - 2017-02-10 ; Livermore, California, United States
Country of Publication:
United States
Language:
English
Subject:
Knight's Landing, GPUs, OpenMP, MPI, OpenACC

Citation Formats

Robey, Robert W., Zamora, Yuliana Yajaira, and Estrada Lupianez, Jenniffer Marie. Experience on New Architectures with the Higrad Code. United States: N. p., 2017. Web.
Robey, Robert W., Zamora, Yuliana Yajaira, & Estrada Lupianez, Jenniffer Marie. Experience on New Architectures with the Higrad Code. United States.
Robey, Robert W., Zamora, Yuliana Yajaira, and Estrada Lupianez, Jenniffer Marie. Tue . "Experience on New Architectures with the Higrad Code". United States. doi:. https://www.osti.gov/servlets/purl/1345120.
@article{osti_1345120,
title = {Experience on New Architectures with the Higrad Code},
author = {Robey, Robert W. and Zamora, Yuliana Yajaira and Estrada Lupianez, Jenniffer Marie},
abstractNote = {},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Tue Feb 21 00:00:00 EST 2017},
month = {Tue Feb 21 00:00:00 EST 2017}
}

Conference:
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:
  • Utilities must gain better control over their resources by determining how energy is actually being used and what caused it to be used at those specific times. With the use of this information, utilities can identify which loads are the easiest to control and which beneficially alter the load curve while incurring the least amount of inconvenience to their customers. This intense lack of information is not specific to any one type of utility. What all gas, water and electric utilities have in common is their same lack of information regarding their customers` consumption patterns. If this were a genericmore » manufacturing company or industry we were examining it would be similar to this organization operating without product sales and marketing information. Acquiring information about one`s customer is a vital criteria to staying in business--especially in a competitive world! Capturing more frequent meter reads and monitoring all customer segments -- commercial, industrial and residential--on a daily and sometimes hourly basis, provides the sales data so vital to addressing the optimization of operations currently being faced by gas, water and electric utilities today! Many forward thinking utilities acknowledge that automation and telecommunication technologies could be harnessed to solve many of these concerns but these same utilities having an in depth knowledge of the status of technology today site two key reasons that stand in the way of installing electronic technology-based initiatives -- end-device costs and communications costs. It is for this reason that multi-utility trials that share the costs and the risks are receiving a serious second look by utilities that are servicing the same customer base. Automated Communication Links has worked on several of these installations (successfully) and would like to take this opportunity to relate some of the findings derived from these actual installations.« less
  • Pipelined processors employ special hardware to prevent the execution of an instruction before all operands are ready or to control the prefetching of instructions when a branch is encountered. An alternative solution to this problem is to ensure that executed code will be free of pipeline hazards. A postpass code reorganizer which guarantees the absence of pipeline hazards in compiled code and exploits delayed branches is described; it has been implemented for the MIPS processor, an experimental VLSI microprocessor. Code reorganization yields an average dynamic improvement of 30percent for one set of benchmark programs. 9 references.
  • A horizontal architecture consists of a number of resources which can operate in parallel and each of which is controlled by a field in the wide instruction word. Such architectures offer the potential for high performance scientific computing at a modest cost. If this potential performance is to be realised, the multiple resources of a horizontal processor must be scheduled effectively. The scheduling task for conventional horizontal processors is quite complex and the construction of highly optimising compilers for them is a difficult and expensive project. The polycyclic architecture is a horizontal architecture with architectural support for the scheduling task.more » The complexity of scheduling conventional horizontal processors and the ease of scheduling polycyclic processors is demonstrated by means of an example. 17 references.« less
  • The scheduling task for conventional horizontal processors is quite complex and the construction of highly optimising compilers for them is a difficult and expensive project. The polycyclic architecture is a horizontal architecture with architectural support for the scheduling task. The cause of the complexity involved in scheduling conventional horizontal processors and the manner in which the polycyclic architecture addresses this issue are outlined. 13 references.
  • Superscalar and Very Long Instruction Word (VLIW) architectures exploit fine-grain parallelism to achieve better performance. Static scheduling techniques, such as trace scheduling and superblock scheduling, can effectively produce compact code for these architectures. In this paper, we present an analytical approach for bookkeeping in code scheduling that alleviates the coding complexity and instruction duplication limitations of the previous approaches. We describe techniques that allow instructions to be moved around loop and if-then-else constructs using global information. We also show that according to the classification of the register sets, certain instructions can be moved around subroutine calls, since their register livemore » ranges can be determined across the procedural boundaries at compile time. Performance is compared with respect to the speed-up, the code size and the scheduling time. Experimental results indicate that the code growth and the speed-up are both improved with small increase in scheduling time.« less