Bibliographic Details
| Title: |
NEANDERTHAL: CLOJURE BIBLIOTEKA ZA BRZE OPERACIJE LINEARNE ALGEBRE. (Croatian) |
| Alternate Title: |
NEANDERTHAL: CLOJURE LIBRARY FOR FAST OPERATIONS OF LINEAR ALGEBRA. (English) |
| Authors: |
Đurić, Dragan |
| Source: |
InfoM; 2015, Issue 54, p31-36, 6p, 1 Chart |
| Abstract (English): |
Neanderthal is a software library that integrates fast native linear algebra libraries with Java platform, targeted at Clojure programming language. The existing pure Java libraries are slow and have arbitrary abstract programming interfaces (APIs), while the libraries that call native libraries have unnecessary overhead in the communication layer, and also have arbitrary APIs. Neanderthal solves these problems by modeling the API on BLAS standard while it minimizes the communication overhead through a JNI-based hand-written layer written in the C programming language. The resulting library has a simpler API that matches the existing literature, while being at least twice faster than the fastest competing solutions. [ABSTRACT FROM AUTHOR] |
| Abstract (Croatian): |
Neanderthal je softverska biblioteka koja integriše brze platformski zavisne biblioteke za linearnu algebru sa Java platformom, namenjena korišćenju u Clojure programskom jeziku. Postojeće čiste Java biblioteke su spore i imaju proizvoljan apstraktni programski interfejs (API), dok biblioteke koje pozivaju platformski zavisne biblioteke imaju nepotrebne gubitke u komunikaciji, i takođe imaju proizvoljan API. Neanderthal ove probleme rešava tako što API modelira po uzoru na BLAS standard dok gubitke u komunikaciji minimizira kroz JNI-bazirani sloj za povezivanje pisan ručno u C programskom jeziku. Rezultujuća biblioteka ima jednostavniji API koji se slaže sa postojećom literaturom, dok je brzina dva i više puta veća u odnosu na postojeća najbrža rešenja. [ABSTRACT FROM AUTHOR] |
|
Copyright of InfoM is the property of Belgrade University, Faculty of Organizational Science and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) |
| Database: |
Complementary Index |
Nájsť tento článok vo Web of Science