Abstract
In this paper, power reduction technique is applied over SRAM and sense amplifier and then calculate the power consumption through cadence tool. Over the conclusion of analysis, A Architecture of cache memory has been done with lowest leakage power. In this paper from different sense amplifiers, we conclude that Charge-Transfer SA have lowest power dissipation, i.e.,11.06 µW. SRAM has 220.078 µW power dissipation after applying leakage power reduction technique such as MTCMOS_technique, Footer Stack Technique, Sleepy Keeper Technique and Sleep-Stack Technique, and there is 98–99% reduction and 75–76% reduction in Charge-Transfer SA. So, after all conclusion, architecture has been made with MTCMOS SRAM memory and MTCMOS Charge-Transfer SA with 98% reduction in power dissipation. This paper describes that MTCMOS_technique applied over different circuits reduces leakage power reduction as well as SRAM and CTSA with MTCMOS_technique in architecture gives low power consumption for a cache memory.
Access provided by Autonomous University of Puebla. Download conference paper PDF
Similar content being viewed by others
Keywords
- MTCMOS CTSA (charge-transfer sense amplifier with MTCMOS_technique)
- MTCMOS SRAM (Static Random Access memory with MTCMOS_technique)
- SA (Sense Amplifier)
1 Introduction
In today’s world, technology needs low power dissipation devices. Due to portable handle devices as they have low number of power plugs at their surrounding, i.e., they need a high battery backup devices [1] which consume less amount of energy during non-working process and as well as in working process.
In proposed work, a design is made under convince of day to day life as it consumes less amount of energy. In this architecture [2], we apply MTCMOS technique over SRAM cell because after applying leakage reduction techniques, we compare all SRAM cell as shown in Table 1 conclusion arises that SRAM with MTCMOS technique consumes lowest power. Similarly, from Table 2, conclusion arises that VMSA, CMSA and CTSA consume lowest power among all sense amplifiers. Due to this reason, leakage reduction techniques [3] are applied over VMSA, CMSA and CTSA. And from Tables 1 and 2, architecture is form of MTCMOS SRAM, WDC, PCH and VMSA, CMSA and CTSA, respectively.
After this analysis, A “MTCMOS CACHE MEMORY ARCHITECTURE” is formed which is made up [4] of MTCMOS SRAM, PCH, WDC and MTCMOS CTSA which shows 98–99% less power consumption [5].
2 MTCMOS Cache Memory Architecture
2.1 Write Driver Circuit
When discharging bl (bit lines) voltage, then WDC reduces the write margin of “SRAM” from highest PCH value [6] as shown in Fig. 1.
2.2 Conventional SRAM
“SRAM” is a 6T design. SRAM works as a cache memory in computers. It is used to store a data. It is made by connection of two invertors back to back with two complementary transistors [7]. The “read” and “write” operations can be done through bit lines in “SRAM” cell as shown in Fig. 2. It is popular due to its high stability property and lowest “static” power dissipation. “Access” transistors which are connected to the bl (bit_lines) are in working process when write line (wl) is enabled so “read” and “write” operation can be done [8]. Figure 3 shows the 8T design, i.e., 6T SRAM cell with 2T technique, i.e., MTCMOS technique [13].
2.3 Sense Amplifier’s
2.3.1 Voltage Mode Sense Amplifier
For sensing the potential difference at bit capacitance, VMSA requires differential discharging [9]. Some voltage differential is developing at bl when WL is enabling. Saen is enabled when sufficient voltage differential is developed in VMSA due to which interconnected inverters follow +ve feedback loop and convert the difference in o/p to full rail o/p [10]. Figure 4 presents the electrical picture of VMSA.
2.3.2 Current-Mode Sense Amplifier
The CMSA manage by sensing the bit_lines current. CMSA does not depend on variation in voltage value developing on bl (bit_lines). It is suitable in reducing the bit line (bl) voltage considering that low voltage at bit line can be clamped at high voltage. Figure 5 presents the electrical diagram of CMSA [11].
2.3.3 Charge-Transfer Sense Amplifier
CTSA works on a principle of charge rearrangement, i.e., from high capacitance bl to low capacitance SA o/p points. Due to this, CTSA consume low power [12]. Figure 6 presents the electrical diagram of CTSA. The circuit is divided into two parts. In first part, CG cascade made with P_1, P_3 and P_5 (and P_2, P_4 and P_6). The pmos P_1 and P_2 are biased at potential Vb. In second part, P_7 through N_11 latches made cross-coupled inverters.
2.3.4 Voltage Latch Sense Amplifier
Figure 7 is a VLSA electrical diagram. P_1, P_2, N_1 and N_2 form the inverters. Differential voltage o/p bit line (bl) converted into full-swing o/p by inverter. The internal node of circuit is charged through bit lines (bl). Difference creates on internal nodes by input bit lines operates the electrical design [13].
2.3.5 Current-Latch Sense Amplifier
CLSA is popular due to low power consumption and an automatic power saving scheme as shown in Fig. 8. In read cycle, small difference on bit line (bl) is a data of cell [14]. The two gates n1 and n2 are connected to bl and blb. The serially connected latch circuit is controlled by current flow of two n-mos.
3 Analysis of Result
Figure 10 represents the block structure of Single Bit MTCMOS CACHE MEMORY ARCHITECTURE [15] implemented with MTCMOS SRAM and MTCMOS CTSA (Fig. 9).
Cache memory architecture made of WDC, PCH Circuit, “SRAM” cell and sense amplifier, i.e., CTSA. Every block is evaluated and described below with their outputs [16].
Figure 10 shows single-bit MTCMOS CACHE MEMORY ARCHITECTURE. The output from WDC is connected to the bit lines of SRAM Cell. O/p is stored in memory cell when write line (i.e., WL = 1) is enabled. Sense amplifier turns on when Saen = 1 [17]. Figure 13 represents the o/p wave form of “MTCMOS CACHE MEMORY ARCHITECTURE”. When bl is charged up to Vdd by PCH circuit, read and write operation can be done. When WE = 1, bit line stored the o/p data, and by charging and discharging the bit lines, write operation can be performed. Now, charge the bit lines up to V_dd (i.e., bl = Vdd), and the SA sensed the change in voltage at o/p points. The o/p matches the data saved in “SRAM” cell, which represent “read 0” and “read 1” operation (Fig. 11).
Table 1 displays the “power dissipation” of SRAM and SRAM with different leakage reduction techniques, whereas Table 2 demonstrates the power dissipation of different sense amplifiers. As today’s need is device with low power consumption, so different leakage reduction techniques are applied over different sense amplifiers as shown in Table 3.
From Table 2, conclusion arises that VMSA, CMSA and CTSA consume low power among all five types of sense amplifier, due to this reason, a Single-Bit Cache Memory Architecture is designed using VMSA, CMSA and CTSA, and their power consumption analysis has been seen as shown in Table 4.
From Table 1, conclusion arises that SRAM with MTCMOS technique consumes low power, and from Table 2, conclusion arises that Charge-Transfer sense amplifier consumes lowest power, and from Table 3, conclusion arises that CTSA with MTCMOS technique consume low power. So, now architecture has been designed having SRAM with MTCMOS technique, PCH, WDC and CTSA with MTCMOS technique, and the complete structure is known as “MTCMOS CACHE MEMORY ARCHITECTURE”. There is 98–99% decrement in power consumption in new design as shown in Table 4 (Table 5).
4 Conclusion
In presented work, MTCMOS CACHE MEMORY ARCHITECTURE has been implemented with MTCMOS SRAM (i.e., SRAM with MTCMOS_technique) and MTCMOS CTSA (i.e., CTSA with MTCMOS_technique). Analysis concludes that MTCMOS CACHE MEMORY ARCHITECTURE with MTCMOS CTSA offers a much superior performance in terms of reduction in power dissipation. It was found that MTCMOS technique implemented in any circuit gives 75–76% reductions in power dissipation and MTCMOS CACHE MEMORY ARCHITECTURE consumes 98–99% less power as compared to conventional architecture. Furthermore, this work has been implemented in array form of MTCMOS CACHE MEMORY ARCHITECTURE.
References
He Y, Zhang J, Wu X, Si X, Zhen S, Zhang B (2019) A half-select disturb-free 11T SRAM cell with built-in write/read-assist scheme for ultralow-voltage operations. IEEE Trans Very Large Scale Integration (VLSI) Syst 27(10):2344–2353. https://doi.org/10.1109/TVLSI.2019.2919104
Pandey S, Yadav S, Nigam K, Sharma D, Kondekar PN (2018) Realization of junctionless TFET-based power efficient 6T SRAM memory cell for internet of things applications. In: Proceedings of first international conference on smart system, innovations and computing. Springer, Singapore, pp 515–523
Jeong H, Oh TW, Song SC, Jung S-O (2018) Sense-amplifier-based flip-flop with transition completion detection for low-voltage operation. IEEE Trans Very Large Scale Integrat (VLSI) Syst
Tao Y, Hu W (2015) Design of sense amplifier in the high speed SRAM. In: International conference on cyber-enabled distributed computing and knowledge discovery, pp 384–387
Shalini AK (2013) Design of high speed and low power sense amplifier for SRAM applications. Int J Sci Eng Res 4(7)
Kaushik CSH, Vanjarlapati RR, Krishna VM, Gautam T, Elamaran V (2014) VLSI design of low power SRAM architectures for FPGAs. In: 2014 international conference on green computing communication and electrical engineering (ICGCCEE), pp 1–4
Nahid R, Singh BP (2013) Static-noise-margin analysis of conventional 6T SRAM cell at 45 nm technology. Int J Comput Appl 66(22)
Saun S, Kumar H (2019) Design and performance analysis of 6T SRAM cell on different CMOS technologies with stability characterization. In: OP conference series: materials science and engineering, vol 561, p 012093
Mohammad B, Dadabhoy P, Lin K, Bassett P (2013) Comparative study of current mode and voltage mode sense amplifier used for 28 nm SRAM. In: 24th international conference on microelectronic, 07 March 2013
Sinha M, Hsu S, Alvandpour A, Burleson W, Krishnamurthy R, Borhr S (2003) High-performance and low-voltage sense-amplifier techniques for sub-90 nm SRAM. In: SOC conference, 2003. Proceedings. IEEE international [Systems-on-Chip]
Dutt R, Abhijeet (2012) High speed current mode sense amplifier for SRAM applications. IOSR J Eng 2:1124–1127
Heller L, Spampinato D, Yao Y, High-sensitivity charge-transfer sense amplifier. In: Solid-state circuits conference. Digest of technical papers. 1975 IEEE international.
Wei Z, Peng X, Wang J, Yin H, Gong N, Novel CMOS SRAM volatge latched sense amplifiers design based on 65 nm technology, pp 3281–3282
Chandankhede RD, Acharya DP, Patra PK (2014) Design of high speed sense amplifier for SRAM. In: IEEE international conference on advanced communication control and computing technologies, pp 340–343
Choudhary R, Padhy S, Rout NK (2011) Enhanced robust architecture of single bit SRAM cell using drowsy cache and super cut-off CMOS concept. Int J Industr Electron Electr Eng 3:63–68
Gajjar JP, Zala AS, Aggarwal SK (2016) Design and analysis of 32 bit SRAM architecture in 90 nm CMOS technology 03(04):2729–2733
Agrawal R, Tomar VK (2018) Analysis of cache (SRAM) memory for core i™ 7 processor. In: 9th international conference on computing, communication and networking technologies (ICCCNT), 2018, p 402
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Agrawal, R., Goyal, V. (2021). Analysis of MTCMOS Cache Memory Architecture for Processor. In: Goyal, V., Gupta, M., Trivedi, A., Kolhe, M.L. (eds) Proceedings of International Conference on Communication and Artificial Intelligence. Lecture Notes in Networks and Systems, vol 192. Springer, Singapore. https://doi.org/10.1007/978-981-33-6546-9_9
Download citation
DOI: https://doi.org/10.1007/978-981-33-6546-9_9
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-33-6545-2
Online ISBN: 978-981-33-6546-9
eBook Packages: EngineeringEngineering (R0)