ram Related Articles

In this article today, we will introduce 9 simple audio amplifier circuit design schematic diagram with detailed explanation, complemented with some knowledge about amplifier.     Catalog   I. What is an Audio Amplifier? II. Nine Simple Audio Amplifier Circuit Design Schematic Diagrams III. Some Knowledge about Amplifier FAQ     I. What is an Audio Amplifier? Let's watch a video first. This video shows us how to make a great sounding LM386 audio amplifier with bass boost An audio amplifier is a device that reconstructs an input audio signal on an output element that produces sound. The reconstructed signal volume and power level are ideal. They are truthful, effective, and low distortion. The audio range is from 20Hz to 20 kHz, so the amplifier must have a good frequency response within this range. Depending on the application, the power varies greatly. From milliwatts of headphones to several watts of TV or PC audio, to dozens of watts of mini home stereo and car audio, to hundreds of watts or more of more powerful household and commercial audio systems, until the power is big enough to meet the sound requirements of an entire cinema or auditorium.   The development of audio amplifiers has gone through three times. They are electron tube (vacuum tube) time, bipolar transistor time, and field-effect transistor time. The electronic tube audio amplifier has a round and sweet timbre, but it has the disadvantages of large volume, high power consumption, unstable operation, and poor high-frequency response. For the bipolar transistor audio amplifier, it has the advantages of a wide frequency band, large dynamic range, high reliability, long life, and high-frequency response is good. However, its static power consumption and on-resistance are very large, and the efficiency is difficult to improve. The third one is the field-effect transistor audio amplifier. It has the same round and sweet timbre as the electron tube and its dynamic range is wide. More importantly, it has a small on-resistance and can achieve high efficiency. Figure 1 II. Nine Simple Audio Amplifier Circuit Design Schematic Diagrams Next, I would like to introduce nine simple audio amplifier circuit design schematic diagrams.   Circuit Diagram 1 This circuit makes full use of the conventional LM317 voltage adjustment chip,  so that it not only completes the voltage stabilization function of the unstabilized voltage after filtering but also realizes the function of amplifying the audio signal picked up by the electret capacitive microphone. The electret capacitive microphone contains an impedance converter based on JFET, which converts the speech signal into a current form and adds it to the RP resistor, causing the corresponding voltage change. 220V AC output 36V unstable DC through transformer and bridge rectifier, and after filtering by the capacitor, the low resistance audio amplification signal input by the LM317 on the DC is fed into the capacitor and output to the loudspeaker. The implementation circuit is shown in figure 2.   Figure 2 After the circuit is installed, the voltage difference between the two inputs of the electret capacitive microphone should be adjusted first. This voltage difference is required to be less than 1.25VDC. By connecting an adjustable resistor between the LM317 adjustment end and the ground, the required limit can be achieved by adjusting the resistance by Rp. Secondly, the audio signal picked up by the microphone is easy to be interfered with by external noise. The addition of C1 can filter out part of the interference signal, but the required signal is also attenuated. Because the internal gain of the LM317 can compensate for the attenuation part, the loss caused by the introduction of C1 is negligible.    In order to avoid excessive loss, the capacity of C1 should be as low as possible, this circuit takes 15F. Finally, it should be noted that the minimum operating current requirement of the LM317 chip is 4 mA when the circuit is working normally, and a load resistor is used to absorb the 4mA current. If a low impedance loudspeaker is used, this load resistance must also be introduced to compensate for the signal distortion. In a practical circuit, if an 8Q impedance loudspeaker is used, at least 420Q load resistance is used to compensate for the possible signal distortion.   Circuit Diagram 2 Figure 3     Circuit Diagram 3 Adjust R1 so that the signal is not distorted at the maximum output, and reduce R2 to output more power. If there is a multimeter, the collector voltage of the transistor can be adjusted to about half of the power supply voltage. Figure 4     Circuit Diagram 4 In this design, the gain control of the preamplifier adopts DC volume control mode is realized as shown in figure 5. The preamplifier is an inverse proportional amplifier composed of a fully differential operational amplifier and resistor. Its gain is determined by the ratio of feedback resistance to input resistance. The external input DC analog control signal Vc is converted into control data through the gain control module (GainCon-troD), which is used to control the ratio of the feedback resistance of the preamplifier to the input resistance, and then adjust the change of the gain. Figure 5   The operational amplifier adopts a two-cascade structure, as shown in figure 6. In the first stage, a folded common-source common-gate amplifier with PMOS input is used to provide a large gain. At the same time, the common-mode range of the input is increased and the flicker noise is reduced. The load of the folded input tube adopts a current source load with a source feedback structure to increase the output impedance and reduce noise. The second stage uses a common-source amplifier to provide a large swing.    In order to maintain the stability of the closed-loop, Miller compensation capacitance is added. At the same time, in order to counteract the influence of the zero points of the right half-plane, the zero adjusting resistance in series with the compensation capacitor is inserted into the feedforward path of the compensation capacitor. In the design of the common-mode feedback circuit, the common-mode feedback structure with resistance distributor and amplifier is adopted.     Figure 6   Circuit Diagram 5 The audio amplifier uses very few peripheral components and works well at 2v. （The circuit is shown in figure 7）   The TDA7052 is a mono amplifier designed for battery-powered portable tape recorders and radios with an internal gain set at 40dB. Now the recorder and radio tend to be miniaturized and the battery consumption is reduced, which means that the power supply voltage is reduced, and the output power is also reduced. In order to compensate for this loss, TDA7052 uses the bridge drive load (ETL) principle, which can make the output power of 8 EU load up to 1.2 w.   Figure 7 lists the working parameters of TDA7052. Except for special instructions, the power supply is 6 v, the load impedance is 80, the input signal frequency is 1 kHz and the ambient temperature is 25 degrees.   Figure 7   Circuit Diagram 6   TDA2822 Fabrication of microphone Power Amplifier Circuit The circuit has few peripheral components, simple fabrication, but surprisingly good sound quality. A dual audio amplifier integrated circuit is used and its main characteristics are high efficiency and low power consumption. The typical value of static working current is only about 6mA. The integrated circuit has strong voltage adaptability (from1.8V to 15V DC) and will still have a power output of about 100mW even if it is used at a low voltage of 1.8V. The specific circuit is shown in figure 8. Figure 8   The electret microphone MIC converts the picked sound signal into an electrical signal, which is introduced from the foot 2 of U1 by C2 and W, and amplified by U1 audio to promote the loudspeaker pronunciation. The machine is connected with a BTL output circuit, which is good for improving sound quality and reducing distortion. At the same time, the output power is increased fourfold. When a 3v voltage is used, the output power is 350mW.   The resistance R1 and R2 are 1/4W metal film resistance, W is a small carbon film potentiometer and C2 is preferably a monolithic capacitor. If there is no good quality ceramic capacitance, select high quality, voltage resistant, and low leakage current electrolytic capacitor for C1, C4, and C3. Choose high sensitivity electret microphone as MIC, choose small button switch or toggle switch for K and choose TDA2822M or TDA2822 or D2822 for U1. According to the numerical value in figure 7, it can work normally without debugging.     Electret microphone detection For example, the R*100 of MF 47 multimeter is used to measure the Great Wall CZ Ⅲ electret microphone. When the black watch pen is connected to the core line and shell of the electret microphone, the multimeter pointer refers to the value at 3k Ω. When blowing hard, the pointer refers to the value at 4k Ω (and the resistance value of some microphones becomes smaller). If you blowhard and the multimeter pointer wobbles very little, you can adjust the two watch pens and try again. If the multimeter needle is still wobbling very little, the electret microphone is damaged.   In application, the drain D of the electret microphone must be connected to the positive electrode of the power supply through a resistance of 4.7 to 10k Ω, and then connected to the amplifier circuit, as shown in figure 9. Figure 9   Circuit Diagram 7    Add an amplifying circuit to the microphone  The electronic components are as follows：                                                                    resistance R1：1k Ω resistance R2：1m Ω resistance R3：1k Ω transistor vT：9014 capacitance：4.7 UF capacitance C2 ：4.7 UF battery：AA size battery     Figure 10    Working principle of amplifier circuit  Fig. 10 is a circuit diagram of the entire microphone amplifier circuit. As you can see from fig. 10, there are only six or seven originals of the whole circuit. The following is a brief description of how it works, in which the resistor R1 is responsible for providing the operating voltage to the microphone, R2 and R3 are responsible for providing the bias voltage for the transistor, and the capacitor C1 is responsible for coupling the signal of the microphone to the transistor for amplification. Finally, the amplified signal is coupled through capacitance C2 and sent back to the positive pole of the microphone line, that is, the outermost shielding layer of the microphone line (that is, the outer layer of copper mesh). Figure 10 is the material or electronic component we use to make it.    Considerations in production  The specifications of the electronic components required for the whole amplifier circuit are as follows: resistance R1：1K Ω resistance R2 ：1m Ω resistance R3 ： 1K Ω transistor VT ：9014 capacitance C1：4.7 μ F capacitance C2 ：4.7 μ F battery： a general AA size battery. △ Generally speaking, it can be used for about half a year if it is used normally.    Pay attention to the following points in the production process： 1. The pin of the transistor must be connected correctly; otherwise, it will not play the role of amplification. The pin distinguishes the following transistors lead down and the flat side facing itself. The transistors are E (emitter), B (base), and C (collector);   2. The microphone head is also polar. (see figure 4 for a specific distinction);   3. The polarity of the coupling capacitance can be distinguished by marking, and the pin with an arrow and marked "-" is a negative electrode, and the positive electrode is generally not marked   Because the components are few or can be directly welded in the shed, the circuit board can be directly installed into the base of the microphone, and the power lead of the circuit board can be connected to the battery slot reserved by the microphone.     Effect Test After trial, the effective distance of the microphone can reach 5 to 6 meters, and the effect is also obvious with the voice input function of Office Word 2003, and the speech can also be accurately recognized about 1 meter away from the microphone.   Circuit Diagram 8 A transistor is required. First, output the MP3 signal and use a lower power tube to amplify it. Then push medium power tube. This can achieve small distortion and have the coupling undone. This transistor circuit is simple, practical, and also easy to make.     Figure 11 The circuit is powered by a 9V single power supply. The input signal is coupled to the base of 9014 through 47uF capacitance. 9014 is responsible for preamplifier, and works in Class A state. 5.6K and 1.5K resistors are bias resistors of 9014, and 5.6K resistors are negative feedback resistors at the same time. 22 Ω resistors is a current series negative feedback resistor, which is used to increase the input impedance and reduce the linear distortion by 9014. The 470 Ω resistor is a 9014 collector load resistor used to convert the 9014 amplified current into a voltage, and two 1N4148 diodes are used to set the post-stage complementary tube in the pre-conduction region. OTL complementary output circuit is composed of 8050 and 8550. 3.3 Ω resistors are negative feedback resistors in series with emitters, which act the same as 22 Ω resistors. The 1000uf capacitor is the output capacitor, which is used to separate the DC and allow the AC signal to pass through. 8050 and 8550 are used as power output tubes to form a complementary push-pull output circuit, and the amplified current of 9014 is further amplified to push the loudspeaker. The static bias current of the push-pull circuit is set by two 1N4148, and the two 1N4148 are temperature compensation elements of two power output transistors at the same time. The voltage at the positive contact of the 1000uF electrolytic capacitor shall be half of the supply voltage. Because the conduction voltage of the silicon transistor base is 0.7 V, the base voltage of 8050 can be obtained to be about 5.2 V. From this, the static bias current of 9014 is (9- 5.2) / 470 =8 [mA]. The emitter voltage of 9014 is 8*22=0.176 V, and the base voltage is 0.176+0.7≈0.87V. Circuit Diagram 9 As long as R3 is increased, the gain of the amplifier can be increased. The parallel capacitor C4 at both ends of R3 is used to provide low resistance path filtering to high frequency to prevent high frequency self-excitation. J1 is a jumper, when J1 is turned on, foot 1 is grounded and the full power amplifier works; when J1 is disconnected, foot 1 is VDD, micro-power off, and the amplifier does not work. Jumper J2 can also control the operation of the amplifier. When J2 is disconnected, the + IN end is unbiased and the amplifier does not work. But if it is connected,  the amplifier works. LM4819 high gain audio amplifier circuit is shown in figure 12. Figure 12   III. Some Knowledge about Amplifier      Distortion  Also referred to as THD+N, Total Harmonic Distortion + Noise is simply a measure of the effect that an amplifier will have on sound output. The lower the distortion is, the closer your amp’s output will be to the original recording’s sound. The more distortion that there is, the more coloration there will be to the sound. Just keep in mind that your speakers will also have an impact upon sound, so choose them wisely by matching them with the right amplifier for the clearest sound.    Left and Right Signals  Crosstalk is a term that refers to the measure of how much of the right signal is mixed with the left signal. Amps come as a single unit, but they need to send signals out separately to the speakers so that you can hear things like a piano on the right and a singer to the left. If there is a lot of crosstalk, though, it will be much more difficult to decipher where the different sounds are coming from.    Power  When you look at an amp’s specs, you will also notice that there is a number for the power output, which is basically how loud the music can go. For the average listener, a 10W amp would be sufficient, as it will let you play your music loudly without creating any distortion. If you are really looking for a super loud amp, though, you can go as high as 100W. It really depends upon what your preferences are, what you will be using your amp for, what speakers you have, and how much room you have.    Connections  Your amp should have plenty of inputs for anything and everything that you wish to plug into it. You could have a 3.5mm connection for your iPod, and you could have a USB connection for your laptop, as a couple of examples. Just don’t sacrifice sound quality for more inputs.    Signal vs. Noise  There will always be some background noise within your amplifier, just as there is always some background noise in your own environment. What you want is an amp that will ensure the background noise is not obvious or perceptible. This will ensure that you will hear all of the music but none of the noise. Checking the signal to noise ratio on an amp will give you clearer insight into how well the product will work in this area.   FAQ   1. What does an audio amplifier do? An audio power amplifier (or power amp) is an electronic amplifier that amplifies low-power electronic audio signals such as the signal from radio receiver or electric guitar pickup to a level that is high enough for driving loudspeakers or headphones.   2. What is the difference between a speaker and an amplifier? Speakers are those things that make sound. The amplifier is what delivers sound to the speakers. Amps are usually radios and speakers are what you plug into the amp/receiver to hear the sound. ... The speakers plug into the sound card which in this case would loosly be called the amplifier.   3. Does amplifier improve sound quality? An amplifier simply increases(magnifies)the components of sound quality. If the quality of the input sound is poor, it will be a louder poor sound ; meaning you will hear the poorness of the sound more. It amplifies everything, the good and the bad.   4. Why do you need an amplifier? An amplifier is the device that turns the low voltage signals from your source equipment into a signal with enough gain to be used to power a pair of speakers. ... The second does the 'heavy lifting' and adds the gain to the signals in order to be used to power a pair of speakers. This is the power amplifier.   5. Which is better amplifier or receiver? A receiver is definitely the more convenient choice of the two, but that doesn't mean that it comes without any downsides. Usually a Lower Quality Amplifier - Though the quality of receiver amps is definitely increasing, you still don't have a completely dedicated amp with a receiver.   6. Which is more important speaker or amplifier? A speakers performance is highly variable and its sound will depend on the amp driving it. But the greatest amp (whatever that is) will sound like crap if the speaker sounds like crap. The quality of the speaker is the ultimate limitation of your system (assuming proper set up and room integration)   7. Can I hook up 8 ohm speakers to a 4 ohm amplifier? Yes, you can use 8 ohm speakers with a 4 ohm amplifier. Just wire two 8 ohm speakers of the same wattage in parallel.   8. How much money should I spend on an amp? If it's just for practicing in your bedroom, you can get a perfectly adequate little practice amp for under $200. If you'll be playing in a band or gigging out, yeah, you probably should expect to spend in the $500-700 range at least.   9. Do you need an amp for a subwoofer? Subwoofers are designed to increase the bass frequencies, resulting in a deep, thumping sound. In most cases, they are paired with an amplifier to boost the sound. If you do not have the funds for both components, you can still hook up a subwoofer without an amplifier; it simply involves a little more know-how.   10. Which transistor is used in amplifier? In most of the electronic circuits, we use commonly NPN transistor configuration which is known as NPN transistor amplifier circuit. Let us consider a voltage divider biasing circuit which is commonly known as a single stage transistor amplifier circuit.     Reference Component LM3886TF LM4652TA LM4765T  

A low-cost, low-risk memory solution has been announced by Microchip, it offers unlimited endurance and safe data storage at power loss. This I2C EERAM memory is an easy to implement, Non-Volatile SRAM (NVSRAM) that can be used by applications that need to constantly or instantaneously record, update or monitor data in sectors which include metering, automotive and industrial.EERAM is a standalone SRAM with shadow EEPROM back-up on a single chip that helps to automatically retain the contents of the SRAM memory when system power is lost. The EERAM offers instant random writes to the array with no write-cycle delay. The I2C EERAM family is available in 4 and 16Kb densities and in standard 8-pin SOIC, TSSOP and PDIP packages. EERAM is available in 3.0 and 5.0V options and in industrial and automotive temperature ranges, of -40 to 85°C and -40 to 125°C respectively and is also available as an automotive-grade memory.Comprised of two familiar and reliable memory technologies on a single chip, EEPROM and SRAM, EERAM offers a robust and dependable data solution that is also the lowest cost non-volatile SRAM solution. EERAM does not require an external battery to safely store data during a power-loss event. Instead, a small, external capacitor is used to provide the energy needed to store the contents of the SRAM on to the EEPROM when system power is lost.Key features:I2C EERAM is a low-cost NVSRAM that requires no external battery to retain dataCombines SRAM with EEPROM back-up on a single chip for lowest-cost NVSRAM solutionEnables instant random writes to the array with no write-cycle delayBenefits applications which constantly or instantaneously record, update or monitor dataAvailable in industrial and automotive temperature rangesReference:KY32-BQ2201SN-NKY32-DP8421AV-20KY32-DS1321KY32-MXD1210CSA+ 

Last March, the AI program AlphaGo beat Korean Go champion LEE Se-Dol at the Asian board game. "The game was quite tight, but AlphaGo used 1200 CPUs and 56,000 watts per hour, while Lee used only 20 W. If a hardware that mimics the human brain structure is developed, we can operate artificial intelligence with less power," points out Professor YU Woo Jong.In collaboration with Sungkyunkwan University, researchers from the Center for Integrated Nanostructure Physics within the Institute for Basic Science (IBS), have devised a new memory device inspired by the neuron connections of the human brain.The research, published in Nature Communications, highlights the devise's highly reliable performance, long retention time and endurance. Moreover, its stretchability and flexibility makes it a promising tool for next-gen soft electronics attached to clothes or body.The brain is able to learn and memorise thanks to a huge number of connections between neurons. The information you memorise is transmitted through synapses from one neuron to the next as an electro-chemical signal.Inspired by these connections, IBS scientists constructed a memory called two-terminal tunnelling random access memory (TRAM), where two electrodes, referred to as drain and source, resemble the two communicating neurons of the synapse.While mainstream mobile electronics, like digital cameras and mobile phones use the so-called three-terminal flash memory, the advantage of two-terminal memories like TRAM is that two-terminal memories do not need a thick and rigid oxide layer."Flash memory is still more reliable and has better performance, but TRAM is more flexible and can be scalable," explains Professor Yu.TRAM is made up of a stack of one-atom-thick or a few atom-thick 2D crystal layers: One layer of the semiconductor molybdenum disulfide (MoS2) with two electrodes (drain and source), an insulating layer of hexagonal boron nitride (h-BN) and a graphene layer.In simple terms, memory is created (logical-0), read and erased (logical-1) by the flowing of charges through these layers. TRAM stores data by keeping electrons on its graphene layer. By applying different voltages between the electrodes, electrons flow from the drain to the graphene layer tunnelling through the insulating h-BN layer.The graphene layer becomes negatively charged and memory is written and stored and vice versa, when positive charges are introduced in the graphene layer, memory is erased.IBS scientists carefully selected the thickness of the insulating h-BN layer as they found that a thickness of 7.5 nm allows the electrons to tunnel from the drain electrode to the graphene layer without leakages and without losing flexibility.Flexibility and stretchability are indeed two key features of TRAM. When TRAM was fabricated on flexible plastic (PET) and stretachable silicone materials (PDMS), it could be strained up to 0.5% and 20%, respectively.In the future, TRAM can be useful to save data from flexible or wearable smartphones, eye cameras, smart surgical gloves, and body-attachable biomedical devices.Last but not least, TRAM has better performance than other types of two-terminal memories known as phase-change random-access memory (PRAM) and resistive random-access memory (RRAM).Reference:MT16JTF51264AZ-1G6M1MD2202-D192MD2203-D576

Fujitsu Semiconductor Limited today announced the launch of the 4 Mbit ReRAM MB85AS4MT, the world's largest density mass-produced ReRAM product. This is the first ReRAM product to be jointly developed with Panasonic Semiconductor Solutions Co., Ltd.The MB85AS4MT is an SPI-interface ReRAM product that operates with a wide range of power supply voltage, from 1.65V to 3.6V. It features an extremely small average current in read operations of 0.2mA at a maximum operating frequency of 5MHz.It is optimal for battery operated wearable devices and medical devices such as hearing aids, which require high density, low power consumption electronic components.Up to this point, Fujitsu Semiconductor has contributed to resolving issues for clients with a need for specifications with greater performance than conventional non-volatile memory, such as EEPROM and serial flash memory, by providing FRAM products, which have such features as high read/write endurance and low power consumption. By adding the new 4 Mbit ReRAM MB85AS4MT to its lineup, Fujitsu Semiconductor can now further expand the options it offers to meet the diversifying needs of its customers.This product features the ability to operate with a wide range of power supply voltage, from 1.65V to 3.6V, can be operated at a maximum of 5MHz through an SPI interface, and uses extremely small average current during read operations (0.2mA operating at 5MHz). It offers the industry's lowest power consumption for read operations in non-volatile memory.The package is a 209mil 8 pin small outline package (SOP), pin-compatible with other non-volatile memory products such as EEPROM. Fujitsu Semiconductor has mounted a 4 Mbit memory density, exceeding the maximum density of serial interface EEPROM, in a miniature 8-pin SOP package size.Fujitsu Semiconductor expects that the MB85AS4MT, featuring high density and low power consumption, will be used in battery-operated wearable devices, medical devices such as hearing aids, and IoT devices such as meters and sensors.Going forward, Fujitsu Semiconductor will continue to provide products and solutions aimed at improving the value and convenience of customers' applications.Reference:GP1S036PKGS-00GXP1-RRB-3R0232-50PKGS-25SXAP1-R 

Random-access memory, or RAM, is where computers like to store the data they're working on. A processor can retrieve data from RAM tens of thousands of times more rapidly than it can from the computer's disk drive.But in the age of big data, data sets are often much too large to fit in a single computer's RAM. The data describing a single human genome would take up the RAM of somewhere between 40 and 100 typical computers.Flash memory—the type of memory used by most portable devices—could provide an alternative to conventional RAM for big-data applications. It's about a tenth as expensive, and it consumes about a tenth as much power.The problem is that it's also a tenth as fast. But at the International Symposium on Computer Architecture in June, MIT researchers presented a new system that, for several common big-data applications, should make servers using flash memory as efficient as those using conventional RAM, while preserving their power and cost savings.The researchers also presented experimental evidence showing that, if the servers executing a distributed computation have to go to disk for data even 5 percent of the time, their performance falls to a level that's comparable with flash, anyway.In other words, even without the researchers' new techniques for accelerating data retrieval from flash memory, 40 servers with 10 terabytes' worth of RAM couldn't handle a 10.5-terabyte computation any better than 20 servers with 20 terabytes' worth of flash memory, which would consume only a fraction as much power."This is not a replacement for DRAM [dynamic RAM] or anything like that," says Arvind, the Johnson Professor of Computer Science and Engineering at MIT, whose group performed the new work. "But there may be many applications that can take advantage of this new style of architecture. Which companies recognize: Everybody's experimenting with different aspects of flash. We're just trying to establish another point in the design space."Joining Arvind on the new paper are Sang Woo Jun and Ming Liu, MIT graduate students in computer science and engineering and joint first authors; their fellow grad student Shuotao Xu; Sungjin Lee, a postdoc in Arvind's group; Myron King and Jamey Hicks, who did their PhDs with Arvind and were researchers at Quanta Computer when the new system was developed; and one of their colleagues from Quanta, John Ankcorn—who is also an MIT alumnus.Outsourced computationThe researchers were able to make a network of flash-based servers competitive with a network of RAM-based servers by moving a little computational power off of the servers and onto the chips that control the USB flash drives. By preprocessing some of the data on the flash drives before passing it back to the servers, those chips can make distributed computation much more efficient. And since the preprocessing algorithms are wired into the chips, they dispense with the computational overhead associated with running an operating system, maintaining a file system, and the like.With hardware contributed by some of their sponsors—Quanta, Samsung, and Xilinx—the researchers built a prototype network of 20 servers. Each server was connected to a field-programmable gate array, or FPGA, a kind of chip that can be reprogrammed to mimic different types of electrical circuits. Each FPGA, in turn, was connected to two half-terabyte—or 500-gigabyte—flash chips and to the two FPGAs nearest it in the server rack.Because the FPGAs were connected to each other, they created a very fast network that allowed any server to retrieve data from any flash drive. They also controlled the flash drives, which is no simple task: The controllers that come with modern commercial flash drives have as many as eight different processors and a gigabyte of working memory.Finally, the FPGAs also executed the algorithms that preprocessed the data stored on the flash drives. The researchers tested three such algorithms, geared to three popular big-data applications. One is image search, or trying to find matches for a sample image in a huge database. Another is an implementation of Google's PageRank algorithm, which assesses the importance of different Web pages that meet the same search criteria. And the third is an application called Memcached, which big, database-driven websites use to store frequently accessed information.Chameleon clustersFPGAs are about one-tenth as fast as purpose-built chips with hardwired circuits, but they're much faster than central processing units using software to perform the same computations. Ordinarily, either they're used to prototype new designs, or they're used in niche products whose sales volumes are too small to warrant the high cost of manufacturing purpose-built chips.But the MIT and Quanta researchers' design suggests a new use for FPGAs: A host of applications could benefit from accelerators like the three the researchers designed. And since FPGAs are reprogrammable, they could be loaded with different accelerators, depending on the application. That could lead to distributed processing systems that lose little versatility while providing major savings in energy and cost."Many big-data applications require real-time or fast responses," says Jihong Kim, a professor of computer science and engineering at Seoul National University. "For such applications, BlueDBM"—the MIT and Quanta researchers' system—"is an appealing solution."  

Silicon Valley startup Flex Logix Technologies has announced that it is now in the business of selling newly developed programmable chip technology to hardware makers—they believe they have found a new way to produce field-programmable gate arrays (FPGAs) that is both cheaper and more amendable to being added to existing systems.For many years hardware and software have held on to a rigid standard—hardware was designed in a very generic way then set firm—the IC chips were given the ability to do certain things, and that was it. Software was then created to run on the platform that had been developed. But this old-school design strategy has one serious flaw—hardware updates take too long and cost too much money. Into this void steps Geoff Tate, CEO of Flex Logix—he claims that engineers with his company have come up with a new and better way to create FPGAs, paving the way for their use in a variety of specialized applications. FPGAs are identical in most respects to regular old microprocessors, at least in how they function—the difference is that they can be programmed. The reason that all of the microprocessors in our phones, computers, etc. aren't FPGAs is because they cost more to make and consume more energy. Tate claims that his engineers have figured out a way to put all the programmable logic on the same chip, something that other big-name makers have not been able to do. That makes things easier for chip designers and for those on the manufacturing end, making the new chips a viable option for certain applications.Tate says that communications gear is one of the best candidates for the new chip design—standards, needs and new advances mean they could benefit greatly from chips that could be updated when needed, instead of going back to the design table every round. Another area where the chips could prove useful is large data centers—just last year Microsoft announced that it had put FPGAs into its Bing search system and saw a speed-up of 95 percent.Instead of making the chips, Flex Logix plans to license its technology to others in the field or directly to those already making the old-fashioned kind. As a sign of the company's optimism, Tate recently told the media that he expects to see products based on the new technology hitting the market as early as next year.