Abstract- In this paper a dual-mode Power/ Performance optimized mode is presented. This dual-operating mode SRAM compensates the delay in Speed-Optimized mode by powering down the pre-charge voltage in the cases the sense amplifiers are slower than designed and a same approach for Power-Optimized mode is applied in which the access time is increased in each iteration. The proposed scheme with a macro cell representing a column with 512 cells is implemented in 0.18um CMOS technology. 45% improvement in read frequency is confirmed in power optimized mode and also a 2× improvement in the failure rate is achieved in comparison to the conventional schemes using the simulation and measurement results.
I. INTRODUCTION
Embedded SRAMs are hugely important components of modern chips. They are used in caches, register files, FIFOs, etc. The growing speed of CPU and DSP reinforces the need to design a more high performance SRAM. On the other hand handheld products such as PDA and cellular phones must very aggressively conserve power. Balancing the tradeoffs between small area, low power and fast reads/writes is an essential part of any SRAM design optimization.
Lower Power Consumption is achieved through performance degradation and higher performance is associated with high power consumption. However in this paper depending on CPU work load and the power source, whether plugged in or battery connected, power or performance is optimized independently.
The read access time is strongly related to the Sense Amplifier, one of the most critical circuits in the periphery of a memory. The use of voltage-mode sense amplifiers causes a speed limitation due to high bit-lines capacitance. Current mode sense amplifiers, on the other hand can great...
... middle of paper ...
..., IEEE Journal of , vol.33, no.8, pp.1208-1219, Aug 1998.
[10]. Houle, R.; , "Simple Statistical Analysis Techniques to Determine Minimum Sense Amp Set Times," Custom Integrated Circuits Conference, 2007. CICC '07. IEEE , vol., no., pp.37-40, 16-19 Sept. 2007
[11]. Aitken, R.; Idgunji, S.; , "Worst-Case Design and Margin for Embedded SRAM," Design, Automation & Test in Europe Conference & Exhibition, 2007. DATE '07 , vol., no., pp.1-6, 16-20 April 2007.
[12]. Sharifkhani, M.; Sachdev, M.; , "SRAM Cell Stability: A Dynamic Perspective," Solid-State Circuits, IEEE Journal of , vol.44, no.2, pp.609-619, Feb. 2009
[13]. Attarzadeh, H.; SharifKhani, M.; Jahinuzzaman, S.M.; , "A scalable offset-cancelled current/voltage sense amplifier," Circuits and Systems (ISCAS), Proceedings of 2010 IEEE International Symposium on , vol., no., pp.3853-3856, May 30 2010-June 2 2010
...hborhood, additive CA are ideally suited for V LSI implementation. Different applications ranging from V LSI test domains to the design of a hardwired version of different CA based schemes have been proposed.
...t, context switch and effects of various history register lengths- due to the increasing interference in the branch history the IHRT scheme performs the best and according to the simulations, it is evident that increasing the history register length often improves the prediction accuracy and effect of context switch- uses the BHT to keep track of branch history, the table needs to be flushed during a context switch are portrayed. On comparison, best predictors gave 94.4% accuracy whereas two-level predictors pushed it to 97%, the other 3% was still substantial miss rate and current version of simple scalar gives 98.8% hit rate.
The EEPROM chip can store up to one kilobytes of data and is divided into 64 words with 16 bits each. Some memory is inaccessible or reserved for later us...
Throughout its history, Intel has centered its strategy on the tenets of technological leadership and innovation (Burgelman, 1994). Intel established its reputation for taking calculated risks early on in 1969 by pioneering the metal-oxide semiconductor (MOS) processing technology. This new process technology enabled Intel to increase the number of circuits while simultaneously being able to reduce the cost-per-bit by tenfold. In 1970, Intel once again led the way with the introduction of the world’s first DRAM. While other companies had designed functioning DRAMs, they had failed to develop a process technology that would allow manufacturing of the devices to be commercially viable. By 1972, unit sales for the 1103, Intel’s original DRAM, had accounted for over 90% of the company’s $23.4 million revenue (Cogan & Burgelman, 2004).
“Which is better, AMD or Intel?” is a question that is constantly debated among people involved with computers. There are many reasons to choose one side over another, as both do have their advantages and disadvantages. Intel and AMD are the most prevalent processor production companies, which in turn creates competition between the two. This question is a by-product of that competition. Only by knowing each company and what their product has to offer, can a person make a decision as to what to buy to suit their needs.
The computer processor is the brain of the system. Our client requires a heavy punch, multi tasking brain. Due to his low income and student status, he does not have a large budget to put towards his computer. The AMD AthlonXP 1900 was chosen because of it's low cost and it's higher performance rating than that of the higher priced equivalent Pentium 4. The AMD AthlonXP 1900's speed is 1.6GHz. This means 1600Hz, which is a very fast processor. Amazingly, the clock speed of this processor is faster, albeit barely, than the Pentium 4 2.0GHz processor, and is much cheaper as well. This is important, as, for less cash, our student can afford a better processor.
Bolla, R., Bruschi, R., Davoli, F., Di Gregorio, L., Donadio, P., Fialho, L., & Szemethy, T. (2013). The green abstraction layer: A standard power-management interface for next-generation network devices. IEEE Internet Computing, 17(2), 82. doi: 10.1109/MIC.2013.39
Basic Mathematics for Electronics seventh edition: Nelson M. Cooke, Herbert F.R Adams, Peter B. Dell, T. Adair Moore; Copyright 1960
This machine has power. Its Fourth-generation quad-core Intel Core i7 processor is recorded to be the fastest performance in the history of MacBook Pro. Thanks to the speeds...
Wolf, D. (2013), Qualcomm: Cometh the Reaper, [Online], Available on: http://siliconhutong.com/category/hardware-and-silicon/, (Accessed on 2 March 2014).
Grundmann, Marius. Physics of Semiconductors: An Introduction Including Devices and Nanophysics. New York: Springer, 2006. Print.
Desktop computers were commonly used since the late 90’s. The increasing number of applications on desktop computers allowed us to do all kinds of different activities like games, music, video, document editing and so on. In comparison with laptop computers, desktop computers have more stable performance, greater capacities, and throughout history, they have proven themselves more reliable to handle every job they have been assigned.
Computers are one of the most popular kinds of electronic devices in the world today. Whether kid or adult, male or female, everyone wants to learn how to operate the computer. People use computers for different purposes such as typing papers, creating websites, making presentations, browsing on the internet, playing games, etc. In fact, many people are still confused about choosing what kind of computer they want to buy. In the world today, there are two types of computers: notebook/laptop and desktop. They both are actually very different in several ways. In this paper, I will compare the size, connectivity, power, and price of notebook and desktop computers. I can make these comparisons because I have both a notebook and a desktop computer in my apartment.
Now, scientists must find new compounds for semiconductors or new methods for their construction and the overall structure of the material. Software engineers have made exciting discoveries in the past decade such as chip stacking, devices operating in the tens of nanometers, and transparent materials for touchscreens and interfaces [2]. Despite these great new ideas they have been limited by poor thermal management. The shrinking size and exponentially growing electrical requirements. The power required for high performance computing applications on some modern processor modules can reach 200–250 W or more which means upt 1 kW of heat per hour or 1895.63 degrees celsius [8]. “Hot spots” on devices are often the determining factor when a device is being judged for reliability as theses “hot spots” can be five or even ten times as hot as the device average [2]. Most of often the these high heat loads have been handled by materials that have a high thermal conductivity so as to spread out the heat from “hot spots”. The more area the heat is spread over the better it can be handled by the device. Besides the downsizing of electronic devices which can condense heat spots, the increase in devices interconnect layers, which serve “as as the streets and highways of the integrated circuit (IC),
The computer has progressed in many ways, but the most important improvement is the speed and operating capabilities. It was only around 6 years ago when a 386 DX2 processor was the fastest and most powerful CPU in the market. This processor could do a plethora of small tasks and still not be working to hard. Around 2-3 years ago, the Pentium came out, paving the way for new and faster computers. Intel was the most proficient in this area and came out with a range of processors from 66 MHz-166 Mhz. These processors are also now starting to become obsolete. Todays computers come equipped with 400-600 Mhz processors that can multi-task at an alarming rate. Intel has just started the release phase of it’s new Pentium III-800MHz processor. Glenn Henry is