#New Technique Could Enable Chips with Thousands of Cores Researchers from MIT have unveiled the first fundamentally new approach to cache coherence in more than three decades,
a memory-management scheme that could help enable chips with thousands of cores. In a modern, multicore chip, every core or processor has its own small memory cache, where it stores frequently used data.
But the chip also has shared a larger cache, which all the cores can access. If one core tries to update data in the shared cache,
other cores working on the same data need to know. So the shared cache keeps a directory
of which cores have copies of which data. That directory takes up a significant chunk of memory:
In a 64-core chip, it might be 12 percent of the shared cache. And that percentage will only increase with the core count.
Envisioned chips with 128,256, or even 1, 000 cores will need a more efficient way of maintaining cache coherence.
At the International Conference on Parallel Architectures and Compilation Techniques in October, MIT researchers unveil the first fundamentally new approach to cache coherence in more than three decades.
Whereas with existing techniques, the directory memory allotment increases in direct proportion to the number of cores, with the new approach, it increases according to the logarithm of the number of cores.
In a 128-core chip that means that the new technique would require only one-third as much memory as its predecessor.
With Intel set to release a 72-core high-performance chip in the near future, that a more than hypothetical advantage.
But with a 256-core chip, the space savings rises to 80 percent, and with a 1, 000-core chip, 96 percent.
When multiple cores are simply reading data stored at the same location, there no problem.
Conflicts arise only when one of the cores needs to update the shared data. With a directory system, the chip looks up
which cores are working on that data and sends them messages invalidating their locally stored copies of it. irectories guarantee that
when a write happens, no stale copies of the data exist, says Xiangyao Yu, an MIT graduate student in electrical engineering and computer science and first author on the new paper. fter this write happens,
no read to the previous version should happen. So this write is ordered after all the previous reads in physical-time order. ime travelwhat Yu and his thesis advisor Srini Devadas,
the Edwin Sibley Webster Professor in MIT Department of Electrical engineering and Computer science realized was that the physical-time order of distributed computations doesn really matter,
core A can keep working away on a piece of data that core B has provided since overwritten that the rest of the system treats core A work as having preceded core B. The ingenuity of Yu
and Devadasapproach is in finding a simple and efficient means of enforcing a global logical-time ordering. hat we do is we just assign time stamps to each operation,
With Yu and Devadassystem, each core has its own counter, and each data item in memory has associated an counter,
When a core reads a piece of data, it takes out a easeon it, meaning that it increments the data item counter to,
As long as the core internal counter doesn exceed 10, its copy of the data is valid. The particular numbers don matter much;
When a core needs to overwrite the data, however, it takes wnershipof it. Other cores can continue working on their locally stored copies of the data,
but if they want to extend their leases, they have to coordinate with the data item owner.
The core that doing the writing increments its internal counter to a value that higher than the last value of the data item counter.
Say, for instance, that cores A through D have all read the same data, setting their internal counters to 1
Core E needs to overwrite the data so it takes ownership of it and sets its internal counter to 11.
Its internal counter now designates it as operating at a later logical time than the other cores:
it will find it owned by core E, to which it sends a message. Core E writes the data back to the shared cache,
and core A reads it, incrementing its internal counter to 11 or higher. Unexplored potentialin addition to saving space in memory
Tardis also eliminates the need to broadcast invalidation messages to all the cores that are sharing a data item.
says Christopher Hughes, a principal engineer at Intel Labs, ut at least to my knowledge, they tend to use physical time.
Simultaneously, the cost of 3-D printers has fallen sufficiently to make them household consumer items.
Now a team of MIT researchers has opened up a new frontier in 3-D printing:
From the discovery of core-forming process for bead-making in ancient Egypt, through the invention of the metal blow pipe during Roman times, to the modern industrial Pilkington process for making large-scale flat glass;
Like other 3-D printers now on the market, the device can print designs created in a computer-assisted design program,
producing a finished product with little human intervention. In the present version molten glass is loaded into a hopper in the top of the device after being gathered from a conventional glassblowing kiln.
far higher than the temperatures used for other 3-D printing. The stream of glowing molten glass from the nozzle resembles honey as it coils onto a platform,
Klein says the printing system is an example of multidisciplinary work facilitated by MIT flexible departmental boundaries in this case
the lab assembled three-dimensional computer models of illared graphene nanostructures, akin to the boron nitride structures modeled in a previous study to analyze heat transfer between layers. his time we were interested in a comprehensive understanding of the elastic and inelastic properties
Rice university, the National Science Foundation (NSF), the National institutes of health and IBM-shared University Research Award supported the research.
The researchers used the NSF-supported DAVINCI supercomputer administered by Rice Ken Kennedy Institute for Information technology n
a core faculty member at Harvard Wyss Institute, has developed now a strategy that has improved experimentally bone repair by boosting the survival rate of transplanted stem cells
and metamaterials offers tantalizing future prospects for technologies such as high resolution optical microscopes and superfast optical computers.
At the macroscale, among other applications, invisibility cloaks could prove useful for 3d displays. This research was funded by the DOE Office of Science a
and metamaterials offers tantalizing future prospects for technologies such as high resolution optical microscopes and superfast optical computers.
At the macroscale, among other applications, invisibility cloaks could prove useful for 3d displays. This research was funded by the DOE Office of Science i
and re-scan it repeating the process until the desired spatial resolution is achieved before combining the data from each scan using a computer algorithm.
Using mouse models, the researchers tested their sunblock against direct ultraviolet rays and their ability to cause sunburn.
The research group already recruited large populations of students to help determine the connectome of a part of the mouse retina
said Conor Walsh, a contributing author of the study, a Wyss Institute core faculty member, an assistant professor of mechanical and biomedical engineering AT SEAS,
and monitors to locate other vehicles on the road enables the Inspiration Trucks to remain in their lane,
As Peter Stone, a University of Texas computer scientist put it, efore it became clear that the technical issues could be addressed,
#A Computer That Can Sniff out Septic Shock Dr. David Hagar treats dozens of patients each day at the intensive care unit at John Hopkins Hospital in Maryland.
A group of computer scientists at John Hopkins University partnered with Hagar, and created an algorithm that can predict septic shock
The computer system sifted through a dataset of over 16,000 patient electronic health records, which includes a historical profile of blood pressure, heart rate,
explains the study lead computer engineer, Suchi Saria. Part of the difficulty is that there may be systematic bias in the medical information recorded.
Saria and her team address this problem by modifying pattern recognition algorithms so the computer can avoid mistaking high-risk patients for low-risk ones.
This computer system can be tailored to many different medical conditions including acute lung injury, pneumonia, and post-rehabilitation illnesses like neuropathy. e are at a very exciting time,
Tan and company built their igh Doppler resolution passive Wi-fi radaron two multi-frequency, software-defined, FPGA-based transceivers (National Instrumentsusrp,
or Universal Software Radio Peripheral. The system compares the reference and surveillance signals, interprets the very small frequency shifts,
so and developed computer models demonstrating that combining graphene with lithium might do the trick. Lithium, they predicted,
In a research paper available on arxiv, the researchers demonstrated in physical experiments that the computer models were indeed correct in their predictions.
Those companies, in turn, typically made it available through smartphone apps or dashboard consoles. The new service, called INRIX Road Weather, adds data gleaned from the actions of the caror instance,
Graphene has long been pursued as a potential replacement for indium tin oxide (ITO) as a transparent electrode material for displays.
Graphene has long been pursued as a potential replacement for indium tin oxide (ITO) as a transparent electrode material for displays.
Right now, Fraunhofer is putting the film on small panels that can be glued to the inside of a car door,
#3-D Printing Software Turns Heart Scans into Surgical Models A new 3-D printing system can transform medical scans of a patient heart into a physical models that help
The efficient system relies on a computer algorithm that requires just a pinch of human guidance to figure out a patient heart structure from MRI scans.
The new software developed by MIT and the Boston Children Hospital, can correctly identify an individual heart anatomical structures by following the lead of a human expert who interprets a small patch equivalent to just one-ninth of the area of each cross section, according to an MIT press release.
and allowing the computer algorithm to infer the rest of the patient heart structure across the rest of the MRI scan 200 cross sections.
The software results were in agreement with human experts interpreting all 200 cross sections 90 percent of the time.
But the new software from a team led by Danielle Pace, an MIT graduate student in electrical engineering and computer science, managed to create a fairly accurate digital 3-D model of each patient heart in just an hour.
The researchers plan to report on their system at the International Conference on Medical Image Computing and Computer Assisted Intervention in October.
They hope to improve the software accuracy by examining patches that appear in several MRI cross sections.
either be based on physical 3-D printed models or virtual 3-D models, with the models based on either human expertise or the computer software.
But separate clinical trials aim to test how a personalized computer model for each individual patient could improve medical care,
in the same way that Microsoft latest Xbox Kinect sensor works. The time of flight camera sends out bursts of microwaves
pediatricians explained how hardware and software specialized for genetic analysis can provide such fast and lifesaving information.
The key piece of technology: A processor from the company Edico Genome that designed to handle the big data of genetics.
Lead researcher Stephen Kingsmore, a pediatrician and genomics expert at Children Mercy Hospital in Kansas city, explains that doctors typically run targeted genetic tests for specific diseases
Using Edico Genome DRAGEN processor, the researchers got this step down from 15 hours to 40 minutes.
After that, Kingsmore team used in-house software to search through the mutations for those associated with a disease that matched the baby symptoms.
noting that Children Mercy is going to make its software packages available as freeware by the end of the year.
The DRAGEN processor delivered its critical speed gains to the hospital servers thanks to its architecture,
He envisions the DRAGEN processor outputting its analysis directly into a patient electronic medical record, where actionable intelligence would be flagged for the physician.
#Bright blue PHOLEDS Almost Ready for TV A new energy-efficient organic LED (OLED) that glows a deep blue is finally close to meeting the most stringent U s. video display brightness requirements,
OLEDS have enabled a new generation of bright, high-quality, low-cost, power-efficient, flexible, lightweight flat panel displays.
Each pixel in an OLED display typically consists of red, green, and blue OLEDS that shine with different brightnesses to produce any desired color.
but developing the kind of bright deep blue PHOLEDS needed for video displays has proven challenging.
work sponsored by Universal Display Corporation and the U s. Air force. The researchers added their new lights nearly meet the most stringent requirements of the National Television systems Committee (NTSC),
#Computer Count of Huge Crowds Now Possible Getting a headcount of crowds numbering in the hundreds of thousands need not strain human eyes any longer.
New software has carried out the first automated crowd count on that scale ever by analyzing aerial photographs of a huge demonstration timesaving innovation that could eventually help save lives
The software, developed by University of Central Florida researchers, can drastically cut the time needed for crowd counts from a week to just half an hour.
and other dangers. utomated computer analysis of such large-scale and dense crowds has never been done before,
said Mubarak Shah, computer science professor and director of the Center for Research in Computer Vision at the University of Central Florida, in a press release.
Shah and his colleagues tested their software on aerial photographs of a demonstration involving thousands of people calling for the independence of the Catalonia province from Spain.
The software analyzed 67 aerial photographs taken of the demonstration in Barcelona and came up with a headcount within 30 minutes.
The images and software calculations were double checked by a team at the Pompeu Fabra University in Spain.
The timesaving software could also make accurate crowd counts a crucial new tool for managing large crowds.
For more, see the IEEE Spectrum article on how the Hajj crowds defy conventional computer simulations.
and a confluence of new hardware technologies that freed computation from the desktop and automated multivariate collection of big data.
Finally, a confluence of hardware and software technology advances are creating opportunities to address this market.
Inexpensive and infinitely configurable mobile devices (enabled by advances in wireless and energy storage) have liberated technology from the office desktop.
inexpensive but sophisticated hardware sensors have emerged to automate the collection of massive data sets. With these technology shifts, exciting technologies like drones, AI, satellite mapping, robotics,
It hardware/software solution enables farmers to analyze every drop of water on their property,
#Apple Patents Learning Computer Vision For Gesture Control Apple has a new patent (via Appleinsider) for 3d gesture control,
specifically describing the tech used to help a computer identify hand motions made by a user.
The key innovation Apple made with the iphone interaction model was based getting touch input right its capacitive screens
All of that leads to a Square business having a core list of emails for customers that is ready to act on.
But square does it without any additional hardware by leveraging the fact that they own the payments stack.
and tracking software to attempt to link those emails to return customers hey can just click a link in the insights dashboard of their Square account.
The setup and execution of a campaign are so simple already that it makes sense to let those sellers using mobile devices primarily to collect campaigns to do it just as easy as those sitting at their desktops.
location and cell network strength to automatically determine whether youe available for a call and display that information to friends.
-and-forget nature of the connection software. Each object connects to the controller app automatically with no pairing
or as an ios app. There are actually two versions of Code Kingdoms: one that free for schools to use,
as part of corporate social responsibility programs when they worked for Intel and ARM. In schools they were using the MIT graphical programming language Scratch,
which involves both DIY hardware and learn to code software, and similarly offers a graphical interface to simplify programming.
But Targett argues Kano is a platform on which the Code Kingdoms product could happily sit.
So, while it giving its software to schools for free it aiming to monetize via the play at-home game version of the product using schools as its low-cost distribution mechanism to get in front of lots of kidseyeballs.
In future it plans to sell premium content for the non-schools version of its software to kidsparents,
#Airware Launches Its Commercial Drone Operating system Drones could save businesses big money by doing dull, dirty,
Commercial businesses can now license Airware Flight Core autopilot technology, Ground Control Station for dispatching drones,
Airware integrated hardware software, and cloud system lets enterprises customize drones to check on gas pipelines,
survey farm land, inspect cell towers, patrol property, or do search & rescue, Now wel see just how many businesses will pay a monthly subscription for Airware to power their drones for industrial inspections, agriculture management, anti-poaching, and more.
But long-term, theyl likely be commoditized as hardware makers around the globe clone and undercut each other on price.
Whatever companies makes the software that runs these drones would make serious profits for years to come.
Following a brief introduction by Airware CEO Jonathan Downey, CTO Buddy Michini demonstrated Airware core technologies:
the Flight Core autopilot system, which provides the brains to each drone that Airware powers,
the Ground Control Station that lets a single user operate a fleet of drones from a Windows laptop or tablet,
and then painstakingly dump the data back to your servers, Airware can handle the whole process.
enhance it with cloud widgets to customize their drone to the day use case, and the data flows back automatically.
After several similar demos, Drone America CEO Mike Richards spoke about using his company larger drones with Airware software for missions outside of traditional industrial uses.
which lets customers create widgets for the Ground Control Station in the C# programming language. These widgets are actually the main reason Airware is sticking with Windows for its tablet app.
Instead of requiring customers to build their own app with the exact drone interface needed, building on Windows will allow them to download the same app to each tablet
but provide different interfaces by simply checking off which widgets should be available to different workers in the field on the cloud backend.
Michini was excited also to talk about the App Core which will provide a full-on Linux computer for developers to work with in addition to the standard Flight Core autopilot and Ground Control Station integration.
When it rolls out to select Airware partners this summer, this new system will let customers take advantage of more advanced sensor
and image processing tools without sacrificing an interface that actually usable for operators in the field.
Airware can roll out new products like App Core without having to find new companies to sell to or upsell its current customers.
It just added investment from Intel capital to the money from GE, A16z, Kleiner, First Round, Google Ventures,
monitor farm land, and survey big tracts of land. Planes and helicopters can be terribly expensive for collecting this overhead video and measurement data, though.
The scanned information is sent then to a remote server in Ozcan laboratory which measures the length of the DNA molecules.
Considering current methods require non-portable heavy machinery and computers, this seems like a godsend for researchers in the field.
a company that develops interactive 3d software and media with a medical focus, to develop an Oculus Rift version of the game.
and allowing them to submit them to a social computing server. A sample app can be found here.
Although Eyewire is continuously making revolutionary strides in the gamification of scientific research, it certainly not the only one doing it.
though building better computer vision AI and thus automating big neurodata analysis our collective, collaborative,
Liquidity core technology was developed and patented out of Stony Brook University. It a membrane with pores of. 2 microns in width, small enough to eliminate microbial contaminants that make up the vast majority of water quality incidents.
Screen Shot 2015-05-04 at 7. 29.15 AM To create the nanofiber thin membrane that could be produced at scale,
Screen Shot 2015-05-04 at 7. 24.11 AM This filtration membrane is highly flexible and durable
Screen Shot 2015-05-04 at 7. 24.34 AM Liquidity tech has been verified third-party by Biovir Laboratories, a microbiological testing facility, that it meets the EPA standard of 99.9999
#Voxiebox Displays 3d Images Just Like R2d2's Message From Princess Leia Voxon, chosen out of the Hardware Alley to do the ildcardpitch during Techcrunch Disrupt NY, demonstrated a truly amazing
Its unique combination of hardware and software, developed over 30 years of tinkering in a New york garage, literally rints lightin three dimensions,
The notorious hacking group called the Chaos Computer Club even replicated the fingerprint of the German Defense Minister.
a full scale immersive 3d environment that enables users to create drawings on hand-held tablets, which can then be manipulated on the tablets to create a 3d design within the space.
As the designers are immersed in their work, for example designing a living room, they can test different furniture options according to the scale
and a dome mirror projecting the image onto the screen. Specialized techniques render the 3d scene onto a spherical projection in real-time In other words,
the tablets act as 3d cursors. Your finger can slide through various objects and select them or you can draw right into the scene.
For now, it not connected to any kinds of software. However the team thinks there are long-term possibilities to build an Internet-of-Things model that could tell consumers how much water
You order on an ipad, your name comes up on a transparent LCD screen box, and you pull out a bowl of quinoa.
Opening today in San francisco at 121 Spear St, Eatsa is a new restaurant designed around technology.
The edict that Microsoft has enterprise on lockdown is dissipating. Huge enterprise player Cisco and Apple announced a ast Lanefor ios enterprise users,
which promises a more streamlined and optimized experience for those enterprise customers using Cisco networks and products.
and ipad. few tweaks will also be put in place to make the hop between a Cisco desk phone
ios is the world best mobile platform, and nearly every Fortune 500 and Global 500 company today has put ios at the center of their mobile strategy. iphone
and ipad have become essential tools for the modern workforce and are changing the way work gets done.
Together with Cisco, we believe we can give businesses the tools to maximize the potential of ios
and help employees become even more productive using the devices they already love. There a lot of words here and not a lot of meat,
because it easy to get your software onto it. Cisco Executive Chairman John Chambers calls it an ngineering
#Mobile devices Will Either Have 3d Sensors Or Suffer Flat Sales When Apple hot on iphone 6ad campaign covered billboards and posters in 24 cities around the world,
and mobile device manufacturers badly want you to know about it. Moves like these are smart in a world where 1. 8 billion digital photos are shared on a daily basis
But Intel new Realsense/Project Tango phone concept shows there is still more valuable data to be captured:
Exploiting depth is the only logical choice mobile device manufacturers have if they are to keep their products competitive.
Breaking 3d Out Of Hollywood Intel isn the only one at the party. Apple recently confirmed acquiring 3d-sensor manufacturer Linx for $20 million.
With depth, a computer can understand the size, shape and distance of all objects within its field of view.
The simplest examples are for image and video editing. An image with depth information would allow you to change things like focus
The IDF15 keynote had Intel CEO Brian Krzanich quickly calculating storage space needed for objects by scanning them with a 3d sensor.
and stick them on a 3d avatar for use in teleconferencing or even video games Further applications exist for 3d printing, design, mapping, object recognition, facial recognition, gesture-based control and more.
Intel Realsense, like Microsoft more famous Kinect cameras, work by projecting a signal (a laser
They work well for indoor situations and benefit from an existing base of software designed for this technique,
Facebook, Google and Microsoft are pouring billions into VR/AR research and acquisitions, and dominate headlines with each new build.
So when you take your first selfie with the next generation of mobile devices, put on your best smile:
Overtext Web Module V3.0 Alpha
Copyright Semantic-Knowledge, 1994-2011