No abstract

Purpose This paper aims to present an effective approach to integrated circuit (IC) throughput enhancement, called TΔT thermal control. It does not require any micro-architectural change of the IC. The only modification is the attachment of an additional temperature sensor at the heatsink boundary. TΔT control technique enables assessment of changes in the dimension of cooling conditions and quick reaction to the dynamic changes in the surrounding environment. As a result, the chip can operate flexibly while minimizing thermal violation. Design/methodology/approach Using additional knowledge about the surroundings, the on-chip temperature is regulated. The approach is first investigated theoretically. To validate the utilized thermal model, the measured temperature values of the designed and fabricated testing device are compared with the simulated one. The authors evaluated the impact of the additional sensor location on the reaction time (RT). Using the Spice model, further investigation helps to verify the hypothesis. Findings The control technique described in this paper showed that the temperature of the chip can be regulated using an additional knowledge of the surrounding environment. It has also been demonstrated that the attachment of an additional temperature sensor close to the cooled surface of the package enables TΔT thermal control technique to react faster (rapid powering up/down of the IC). Therefore, this lowers the risk of shutdown while keeping the temperature close to the thermal limit (the maximal temperature of the chip) for a significant period. The simulation results showed that a higher ambient temperature leads to diminution of the interval in which the on-chip temperature stays almost constant when TΔT technique is used (time shift). Originality/value In this study, a new thermal throttling technique that uses the full physical ability of the chip operating under thermal constraint has been evaluated.

This paper presents a new possibility of clock frequency/voltage control in microsystems i.e. high performance processors, exploiting information about cooling efficiency. In this paper, we propose an approach that better exploits the thermal abilities of a chip fixed to cooling system in order to eliminate its energy accumulation. For the purpose of the proposed method, the calculation of so called time shift (TS) is introduced. TS is defined as the duration where the computational system can perform the task at higher frequency without any thermal violation when the chip temperature is close to critical thermal threshold. The analogy between thermal and electrical parameters allows to model RC thermal compact model of structure (chip fixed to the cooling system). Based on this assumption, the authors compute the TS value versus different parameters using RC thermal compact model in Spice environment. The results indicate that TS could fulfil a significant part of die total working time. As an effect the proposed approach may be a means for increasing average clock frequency or voltage supply, consequently enhancing the system's throughput.

No abstract

No abstract