scholarly journals Thermal Management of On-Chip Caches Through Power Density Minimization

Author(s):  
Ja Chun Ku ◽  
S. Ozdemir ◽  
G. Memik ◽  
Y. Ismail

Author(s):  
Ja Chun Ku ◽  
S. Ozdemir ◽  
G. Memik ◽  
Y. Ismail


Electronics ◽  
2020 ◽  
Vol 9 (2) ◽  
pp. 346 ◽  
Author(s):  
Lili Shen ◽  
Ning Wu ◽  
Gaizhen Yan

By using through-silicon-vias (TSV), three dimension integration technology can stack large memory on the top of cores as a last-level on-chip cache (LLC) to reduce off-chip memory access and enhance system performance. However, the integration of more on-chip caches increases chip power density, which might lead to temperature-related issues in power consumption, reliability, cooling cost, and performance. An effective thermal management scheme is required to ensure the performance and reliability of the system. In this study, a fuzzy-based thermal management scheme (FBTM) is proposed that simultaneously considers cores and stacked caches. The proposed method combines a dynamic cache reconfiguration scheme with a fuzzy-based control policy in a temperature-aware manner. The dynamic cache reconfiguration scheme determines the size of the cache for the processor core according to the application that reaches a substantial amount of power consumption savings. The fuzzy-based control policy is used to change the frequency level of the processor core based on dynamic cache reconfiguration, a process which can further improve the system performance. Experiments show that, compared with other thermal management schemes, the proposed FBTM can achieve, on average, 3 degrees of reduction in temperature and a 41% reduction of leakage energy.



Integration ◽  
2021 ◽  
Author(s):  
Nizar Dahir ◽  
Ammar Karkar ◽  
Maurizio Palesi ◽  
Terrence Mak ◽  
Alex Yakovlev


Author(s):  
Jiashen Li ◽  
◽  
Yun Pan ◽  

The improvement of chip integration leads to the increase of power density of system chips, which leads to the overheating of system chips. When dispatching the power density of system chips, some working modules are selectively closed to avoid all modules on the chip being turned on at the same time and to solve the problem of overheating. Taking 2D grid-on-chip network as the research object, an optimal scheduling algorithm of system-on-chip power density based on network-on-chip (NoC) is proposed. Under the constraints of thermal design power (TDP) and system, dynamic programming algorithm is used to solve the optimal application set throughput allocation from bottom to top by dynamic programming for the number and frequency level of each application configuration processor under the given application set of network-on-chip. On this basis, the simulated annealing algorithm is used to complete the application mapping aiming at heat dissipation effect and communication delay. The open and closed processor layout is determined. After obtaining the layout results, the TDP is adjusted. The maximum TDP constraint is iteratively searched according to the feedback loop of the system over-hot spots, and the power density scheduling performance of the system chip is maximized under this constraint, so as to ensure the system core. At the same time, chip throughput can effectively solve the problem of chip overheating. The experimental results show that the proposed algorithm increases the system chip throughput by about 11%, improves the system throughput loss, and achieves a balance between the system chip power consumption and scheduling time.



Author(s):  
Aleš Chvála ◽  
Robert Szobolovszký ◽  
Jaroslav Kováč ◽  
Martin Florovič ◽  
Juraj Marek ◽  
...  

In this paper, several methods suitable for real time on-chip temperature measurements of power AlGaN/GaN based high-electron mobility transistor (HEMT) grown on SiC substrate are presented. The measurement of temperature distribution on HEMT surface using Raman spectroscopy is presented. We have deployed a temperature measurement approach utilizing electrical I-V characteristics of the neighboring Schottky diode under different dissipated power of the transistor heat source. These methods are verified by measurements with micro thermistors. The results show that these methods have a potential for HEMT analysis in thermal management. The features and limitations of the proposed methods are discussed. The thermal parameters of materials used in the device are extracted from temperature distribution in the structure with the support of 3-D device thermal simulation. The thermal analysis of the multifinger power HEMT is performed. The effects of the structure design and fabrication processes from semiconductor layers, metallization, and packaging up to cooling solutions are investigated. The analysis of thermal behavior can help during design and optimization of power HEMT.



2017 ◽  
Author(s):  
Nenad Miljkovic ◽  
Thomas Foulkes ◽  
Junho Oh ◽  
Patrick Birbarah ◽  
Robert Pilawa-Podgurski ◽  
...  


Author(s):  
Tim Wegner ◽  
Martin Gag ◽  
Dirk Timmermann

With the progress of deep submicron technology, power consumption and temperature-related issues have become dominant factors for chip design. Therefore, very large-scale integrated systems like Systems-on-Chip (SoCs) are exposed to an increasing thermal stress. On the one hand, this necessitates effective mechanisms for thermal management and task mapping. On the other hand, application of according thermal-aware approaches is accompanied by disturbance of system integrity and degradation of system performance. In this chapter, a method to predict and proactively manage the on-chip temperature distribution of systems based on Networks-on-Chip (NoCs) is proposed. Thereby, traditional reactive approaches for thermal management and task mapping can be replaced. This results in shorter response times for the application of management measures and therefore in a reduction of temperature and thermal imbalances and causes less impairment of system performance. The systematic analysis of simulations conducted for NoC sizes up to 4x4 proves that under certain conditions the proactive approach is able to mitigate the negative impact of thermal management on system performance while still improving the on-chip temperature profile. Similar effects can be observed for proactive thermal-aware task mapping at system runtime allowing for the consideration of prospective thermal conditions during the mapping process.



Sign in / Sign up

Export Citation Format

Share Document