A few week ago we bring you news about theHyperwall-2which was build up by NASA and dubbed the “ world ’s highest resolution visualization system . ” However , it appears that that claim has already pass to the Hiperspace , or the “ Highly Interactive Parallelized Display Space project ” developed by UC San Diego . According to the press freeing , the showing features about 287 million pixel of sieve resolution - transcend Hyperwall-2 ’s record by about 10 % . It is also about 30 % bigger than the original Hiperspace Wall developed in 2006 at 31.8 feet wide and 7.5 infantry improbable .
https://gizmodo.com/most-powerful-visualization-system-ever-is-faster-than-5020115
The expanded system feature “ 70 high - resolution Dell 30″ displays , arranged in 14 columns of five displays each . Each ’ roofing tile ’ has a resolution of 2,560 by 1,600 pixels — bringing the combine , seeable resolving to 35,640 by 8,000 pixels , or more than 286.7 million pel in all . ” To power this beast , the system utilizes 18 Dell XPS 710/720 computer with Intel quad - core CPUs and dual Nvidia FX5600 GPUs . All in all we are talking about 100 C.P.U. kernel and 38 GPUs cranking out 20 teraflops of peak processing major power and 10 tebibyte of storage ( which increase due to the on and off campus OptIPuter infrastructure ) . Now , that is a geek sexual climax .
Hiperspace is already being used for a wide-eyed raiment of inquiry applications including seismic activity models , mood - change predictions , the social system of the human brain . The full details are uncommitted in the press acquittance below .
UCSD News Release
July 9 , 2008
UC San Diego Unveils World ’s Highest - Resolution Scientific Display System
Calit2 Also Releases New Version of CGLX Cluster Visualization Framework
As the size of complex scientific data point sets uprise exponentially , so does the need for scientist to research the information visually and collaboratively in ultra - high answer environments . To that end , the California Institute for Telecommunications and Information Technology ( Calit2 ) has unveiled the high-pitched - resolution display system of rules for scientific visualization in the humanity at the University of California , San Diego .
The Highly Interactive Parallelized Display Space ( HIPerSpace ) features nearly 287 million pel of screen resoluteness – more than one dynamic pixel for every U.S. citizen , base on the 2000 Census .
The HIPerSpace is more than 10 percent large ( in terms of pel ) than the second - largest display in the world , constructed of late at the NASA Ames Research Center . That 256 - million - picture element organisation , known as the hyperwall-2 , was developed by the NASA Advanced Supercomputing Division at Ames , with support from Colfax International .
The expanded show at Calit2 is 30 percentage bountiful than the first HIPerSpace wall at UCSD , build in 2006 . That organisation was moved to a larger location in Atkinson Hall , the Calit2 building at UCSD , where it was expanded by 66 million pixels to take reward of the fresh space . The system was used officially for the first time on June 16 to demonstrate applications programme for a delegation from the National Geographic Society .
“ Amazingly it took our team less than a day to deplumate down the original wall , relocate and spread out it , ” say Falko Kuester , principal police detective of the HIPerSpace system . “ The eminent solvent show take us more than half - way to our ultimate goal of build a half - billion - pixel tiled display system to give researcher an unprecedented ability to look loosely at large data sets while also zoom in to the tiniest item . ”
Kuester is the Calit2 Professor of Visualization and Virtual Reality , and associate prof in the Jacobs School of Engineering ’s department of Structural Engineering as well as Computer Science and Engineering . He also lead the Graphics , Visualization and practical Reality Lab ( GRAVITY ) , which is acquire the HIPerSpace engineering science .
Calit2 ’s expanded HIPerSpace is an ultra - scale visualization environment develop on a multi - tile paradigm . The scheme features 70 high - result Dell 30″ displays , dress in fourteen newspaper column of five display each . Each ’ roofing tile ’ has a firmness of 2,560 by 1,600 pel – bringing the combine , visible settlement to 35,640 by 8,000 pixels , or more than 286.7 million picture element in all . “ By using with child , high - resolving power tiles , we also have minimized the amount of space taken up by the frame , or bezels , of each show , ” say Kuester . “ bezel will eventually evaporate , but until then , we can shrink their beguilement by keeping the high potential proportion of screenland area to each tile ’s bezel . ” Including the pixels enshroud behind the bevels of each exhibit , which give the “ Gallic room access ” show , the efficacious total image size is 348 million pel .
At 31.8 feet wide and 7.5 feet tall ( 9.7 m x 2.3 m ) , the HIPerSpace is already being used by a full cooking stove of research groups at UC San Diego , which want to be able-bodied to reckon their largest data set while also drill down to the small constituent on the same screen . A team from the Center of Interdisciplinary Science for Art , Architecture and Archaeology ( CISA3 ) go to Florence to laser - scan the main hall of the Palazzo Vecchio , and the center ’s researchers at Calit2 can now manipulate the computer good example , depicting all 2.5 billion data point points and explore the place in veridical time . Other scientists model the impact of seismic activity on structures , climate - modification foretelling , the social system of the human mind , to name a few such program .
for run simulations and explore data interactively , the developer of the HIPerSpace have build into the environment a large computer and graphics processing bunch . The wall is powered by 18 Dell XPS 710/720 calculator with Intel quad - core fundamental processing units ( central processor ) and three-fold nVIDIA FX5600 artwork processing units ( GPUs ) . A head thickening and six streaming leaf node nail the computer hardware pool for a total of 100 CPU pith and 38 GPUs . Thus the HIPerSpace system offers around 20 teraflop of peak processing power and 10 terabytes of storage , but its access to computer science and storage capacity increase dramatically because the wall is an inbuilt part of the National Science Foundation - funded OptIPuter infrastructure on , and beyond , the UCSD campus , including the so - call “ OptIPortal ” tile show systems ( some with as few as four tiles ) that are the chief goal - point for scientists using the infrastructure .
“ The HIPerSpace is the largest OptIPortal in the worldly concern , ” said Calit2 Director Larry Smarr , a innovator of supercomputing practical app and principal police detective on the OptIPuter project . “ The wall is connected by gamey - performance optical networking to the remote OptIPortals worldwide , as well as all of the compute and storage resources in the OptIPuter base , creating the basis for an OptIPlanet Collaboratory . ”
“ We have full access to the OptIPuter imagination , which drastically increase the CPUs , GPUs and storage at our garbage disposal , ” added Kuester . “ Nodes are interconnect via a dedicated Gb subnet and link up into the OptIPuter fabric with a 10 Gigabits - per - second [ Gbps ] uplink . ”
In addition to 10Gbps connectivity to resources at nine locations on the UCSD campus , including Calit2 and the San Diego Supercomputer Center ( SDSC ) , the OptIPuter provides the HIPerSpace system with up to 2Gbps in dedicated fiber connectivity with its herald HIPerWall at Calit2 on the UC Irvine campus ( and its or so 205 million picture element ) . As a result , scientist can gather at the same time in front of the wall in San Diego and Irvine and explore , break down and collaborate in unison while see material - clip , rendered graphics of large information readiness , video streams and telepresence videoconferencing across nearly half a billion pixels .
HIPerSpace is serving as a visual analytics research blank space with applications in Earth systems skill , alchemy , astrophysics , medicine , forensics , fine art and archaeology , while enabling fundamental piece of work in calculator art , visualisation , networking , data densification , cyclosis and human - computer interaction .
In particular , HIPerSpace is a enquiry testbed for visualisation frameworks require for monolithic resolution digital wallpaper displays of the skinny future that will leverage bezel - barren tiles and allow for continuous ocular content .
departure of CGLX Version 1.2.1
The most notable of these fabric is the Cross - Platform Cluster Graphics Library ( CGLX ) , which introduces a fresh approach to high-pitched - operation hardware quicken visualization on ultra - high - resolve video display system . It provides a clustering direction framework , a development API as well as a take set of cluster - quick lotion . Coinciding with the launch of the expanded HIPerSpace organisation , Calit2 today announced the official liberation of CGLX version 1.2.1 , useable for downloading at http://vis.ucsd.edu/~cglx . “ There is no reason why you require to commence from scrawl every metre you want to program an app for a visual image clustering , ” say CGLX developer Kai - Uwe Doerr , project scientist in Kuester ’s lab . “ CGLX was break to enable everybody to write real - sentence art applications for visualisation clusters . The fabric takes care of networking , event treatment , access to hardware - accelerated rendering , and some other things . Users can focalise on writing their app as if they were compose them for a single desktop . ”
With the emergence of OptIPortal technology , ultra - high resolution multi - tile video display environments are no longer limited to a few prime enquiry facilities with highly specialized inquiry teams digest them . As a result , an nonrational yet hefty development framework is ask that supports fundamental research while enabling expert as well as novice exploiter to utilize these systems . From a high-pitched - level view , CGLX make a distributed , parallel computer graphic context and manages its state and events transparently – allowing the user to focus on content and context rather than how fork over thickening and displays are conflate to show the final visual . CGLX enables OpenGL programme , recrudesce for a exclusive workstation , to be executed on a bombastic - scale tiled visualization grid with minimal or no changes to the original code . The hand out nature of the fabric endorse and boost the development of program to yield visual analytics infrastructures , which enable researcher to collaboratively view , interrogate , correlate and manipulate information in real fourth dimension with ocular resolutions well beyond a unmarried workstation . cardinal features of the model let in :
– Cross - platform , hardware - accelerated interpreting ( UNIX and Mac OSX livelihood ) ;
– Synchronized , multilayer OpenGL context support ;
– Distributed consequence management ; and
– Scalable multi - display support .
app using CGLX include a real - time watcher for gigapixel images and image aggregation , video playback , television streaming , and visualization of multi - dimensional models . The CGLX fabric is already used by nearly all 90 megapixel - plus OptIPortals worldwide , and it is uncommitted for Linux ( Fedora , RedHat , Suse ) , Rocks Cluster Systems ( bundled in the hiperroll ) , and Mac OSX ( leopard , Panthera tigris for ppc and Intel ) . CGLX is so conciliatory that it can even be scale down to run on a commodity laptop computer . “ With CGLX , ” explained Falko Kuester , “ investigator can last center on solving involve visualization and data analysis challenge on next - generation visual analytics cyberinfrastructure . ”
More than 800,000 frames from the Spitzer Space Telescope were stitched together to make this portrait of dust and superstar radiating in the inner Milky Way . An software break for the HIPerSpace rampart allow Calit2 to exhibit this and other large data point solidifying topically while connecting to distant entrepot clusters .
Two researchers in Kuester ’s lab – Kevin Ponto and So Yamaoka – are develop ocular analytics techniques to exhibit gigapixel imagery at interactional ( real - time ) speeds on ultra - high resolution displays , notably the HIPerSpace paries . In a forthcoming issue , Ponto and Yamaoka demonstrate an coating they developed on top of CGLX for use on the HIPerSpace wall . It uses OptIPuter networking to connect to remote warehousing cluster host target data set , include the Spitzer Space Telescope Survey ( for which each look-alike of the interior Milky Way is 24,752 by 13,520 pixels ) , and NASA ’s down Marble visualizations of the Earth at monthly interval ( 86.4 million x 43.2 million pixel ) .
“ These extremist - scale visualization technique load data point adaptively and progressively from meshwork bind storage , requiring only a small local computer memory footprint on each display node , while avoiding data reverberation , ” explained graduate - educatee Ponto . “ All datum is effectively loaded on demand in accordance with the topically available show resources . ” Added fellow Computer Science and Engineering PhD student Yamaoka : “ A render node driving a single four - megapixel show , for example , will only get the data needed to fill that display at any given point in metre . If the consider stance is updated , the needed data is again fetched , on need . ”
Daily Newsletter
Get the good tech , scientific discipline , and civilization news in your inbox day by day .
news program from the future , delivered to your present .
Please select your trust newssheet and take your email to upgrade your inbox .
