Using Multiple Threads to Accelerate Single Thread Performance

Author(s):  
Zehra Sura ◽  
Kevin OBrien ◽  
Jose Brunheroto
2017 ◽  
Vol 207 ◽  
pp. 603-608 ◽  
Author(s):  
Toshinaka Shinbutsu ◽  
Shuichi Amano ◽  
Teruie Takemasu ◽  
Toshihiko Kuwabara ◽  
Jyo Shimura

Author(s):  
Yang Jie ◽  
Li Haitao ◽  
Rui Chengjie ◽  
Wei Wenjun ◽  
Dong Xuezhu

All of the cutting edges on an hourglass worm gear hob have different shapes and spiral angles. If the spiral angles are small, straight flutes are usually adopted. But for the hob with multiple threads, the absolute values of the negative rake angles at one side of the cutting teeth will greatly affect the cutting performance of the hob if straight flutes are still used. Therefore, spiral flutes are usually adopted to solve the problem. However, no method of determination of the spiral flute of the hourglass worm gear hob has been put forward till now. Based on the curved surface generating theory and the hourglass worm forming principle, a generating method for the spiral flute of the planar double enveloping worm gear hob is put forward in this paper. A mathematical model is built to generate the spiral flute. The rake angles of all cutting teeth of the hob are calculated. The laws of the rake angles of the cutting teeth of four hobs with different threads from one to four threads are analyzed when straight flutes and spiral flutes are adopted respectively. The laws between the value of the negative rake angles of the hob with four threads and the milling transmission ratio are studied. The most appropriate milling transmission ratio for generating the spiral flute is obtained. The machining of the spiral flutes is simulated by a virtual manufacturing system and the results verify the correctness of the method.


Solar Physics ◽  
2021 ◽  
Vol 296 (8) ◽  
Author(s):  
J. Threlfall ◽  
J. Reid ◽  
A. W. Hood

AbstractMagnetohydrodynamic (MHD) instabilities allow energy to be released from stressed magnetic fields, commonly modelled in cylindrical flux tubes linking parallel planes, but, more recently, also in curved arcades containing flux tubes with both footpoints in the same photospheric plane. Uncurved cylindrical flux tubes containing multiple individual threads have been shown to be capable of sustaining an MHD avalanche, whereby a single unstable thread can destabilise many. We examine the properties of multi-threaded coronal loops, wherein each thread is created by photospheric driving in a realistic, curved coronal arcade structure (with both footpoints of each thread in the same plane). We use three-dimensional MHD simulations to study the evolution of single- and multi-threaded coronal loops, which become unstable and reconnect, while varying the driving velocity of individual threads. Experiments containing a single thread destabilise in a manner indicative of an ideal MHD instability and consistent with previous examples in the literature. The introduction of additional threads modifies this picture, with aspects of the model geometry and relative driving speeds of individual threads affecting the ability of any thread to destabilise others. In both single- and multi-threaded cases, continuous driving of the remnants of disrupted threads produces secondary, aperiodic bursts of energetic release.


2021 ◽  
Vol 18 (3) ◽  
pp. 1-26
Author(s):  
Mustafa Cavus ◽  
Mohammed Shatnawi ◽  
Resit Sendag ◽  
Augustus K. Uht

Lookup operations for in-memory databases are heavily memory bound, because they often rely on pointer-chasing linked data structure traversals. They also have many branches that are hard-to-predict due to random key lookups. In this study, we show that although cache misses are the primary bottleneck for these applications, without a method for eliminating the branch mispredictions only a small fraction of the performance benefit is achieved through prefetching alone. We propose the Node Tracker (NT), a novel programmable prefetcher/pre-execution unit that is highly effective in exploiting inter key-lookup parallelism to improve single-thread performance. We extend NT with branch outcome streaming (BOS) to reduce branch mispredictions and show that this achieves an extra 3× speedup. Finally, we evaluate the NT as a pre-execution unit and demonstrate that we can further improve the performance in both single- and multi-threaded execution modes. Our results show that, on average, NT improves single-thread performance by 4.1× when used as a prefetcher; 11.9× as a prefetcher with BOS; 14.9× as a pre-execution unit and 18.8× as a pre-execution unit with BOS. Finally, with 24 cores of the latter version, we achieve a speedup of 203× and 11× over the single-core and 24-core baselines, respectively.


2021 ◽  
pp. 107780042110658
Author(s):  
Danah Henriksen ◽  
Edwin Creely ◽  
Rohit Mehta

With the emergence of Western posthuman understandings, new materialism, artificial intelligence (AI), and the growing acknowledgment of Indigenous epistemologies, an ongoing rethinking of existing assumptions and meanings about creativity is needed. The intersection of new technologies and philosophical stances that upend human-centered views of reality suggests that creativity is not an exclusively “human” activity. This opens new possibilities and assemblages for conceiving of creativity, but not without tensions. In this article, we connect multiple threads, to reimagine creativity in light of posthuman understandings and the possibilities for creative emergence beyond the Anthropocene. Creativity is implicated as emerging beyond non-human spaces, such as through digitality and AI or sources in the natural world. This unseats many understandings of creativity as positioned in Euro-Western literature. We offer four areas of concern for interrogating tensions in this area, aiming to open new possibilities for practice, research, and (re)conceptualization beyond Western understandings.


Hand ◽  
2018 ◽  
Vol 15 (4) ◽  
pp. 495-501
Author(s):  
Angel Farinas ◽  
Michael Stephanides ◽  
Steven Schneeberger ◽  
Alonda Pollins ◽  
Nancy Cardwell ◽  
...  

Background: Epitendinous sutures not only join the 2 severed tendon edges but also supply strength and support to the repair. Multiple techniques have been described, but none of them include another thread of suture. This could potentially increase the strength of the repair without affecting gliding. Methods: Caprine tendons were harvested, transected, and sutured with 6-0 Prolene. Four groups were created: single thread running (SR), single thread locking (SL), double thread running (DR), and double thread locking (DL). An Instron 5542 was used to pull the repaired tendons apart, and the energy at the break was calculated (gf/mm). For gliding resistance, we harvested a human A2 pulley. A pre- and postrepair value was obtained, and a ratio was elaborated. A single-factor analysis of variance and independent sample t tests were performed. Results: The SR group had a mean energy at break of 9339.71 ± 1630.212 gf/mm; the SL group, 9629.96 ± 1476.45 gf/mm; and the DR group, 9600.221 ± 976.087 gf/mm, with no statistical significance. The DL group was significantly higher at 14 740.664 ± 2596.586 gf/mm ( P < .05). When comparing SR with DL for gliding, SR had less than half of the resistance than DL (0.018 ± 0.004 and 0.049 ± 0.015 N/mm, respectively), with statistical significance ( P < .05). Conclusion: Using a single suture thread for running epitendinous repair is no different than locking with a single thread or using an extra thread in a running fashion. Performing a double suture thread with a locking technique is significantly stronger than the previously mentioned repairs, with the disadvantage of more resistance at the pulley.


Sign in / Sign up

Export Citation Format

Share Document