IU to acquire fastest university-owned AI supercomputer

Big Red 200 will power advanced machine learning applications using massive data sets and faster scientific calculations in a broad range of fields, including medicine, climate change, addiction and cybersecurity


BLOOMINGTON, Ind., and SEATTLE, Wash. — India­na Uni­ver­si­ty and Cray Inc. have announ­ced IU’s acqui­si­ti­on of the fas­test uni­ver­si­ty-owned super­com­pu­ter in the nati­on to sup­port its advan­ced rese­arch in arti­fi­cial intel­li­gence, machi­ne lear­ning, data ana­ly­tics, and sci­en­ti­fic and medi­cal research.

The new sys­tem, Big Red 200, is named for the IU Bicen­ten­ni­al and will replace the high­ly suc­cess­ful Big Red II Cray super­com­pu­ter, which began ser­vice in 2013 but is beco­m­ing obso­le­te. Big Red 200 will beco­me ful­ly ope­ra­tio­nal on IU’s bicen­ten­ni­al anni­ver­s­a­ry on Jan. 20, 2020.

How fast is Big Red 200? It would take ever­yo­ne in the sta­te of India­na more than 28 years — per­forming one cal­cu­la­ti­on per second 24 hours a day, 7 days a week, 365 days a year — to per­form the same num­ber of cal­cu­la­ti­ons that Big Red 200 can do in just one second. It is almost 300 times fas­ter than the ori­gi­nal Big Red super­com­pu­ter from 15 years ago.

The­re has been a huge growth in the use of AI in near­ly every area of sci­ence in recent years,” IU Vice Pre­si­dent for Rese­arch Fred Cate said. “Com­mon to many of the­se is the app­li­ca­ti­on of sophisti­ca­ted machi­ne lear­ning tech­ni­ques to mas­si­ve amounts of data. Essen­ti­al to this is enor­mous com­pu­ting power spe­cial­ly con­fi­gu­red to opti­mi­ze the­se kinds of cal­cu­la­ti­ons. Big Red 200 will bring this natio­nal­ly com­pe­ti­ti­ve capa­bi­li­ty to India­na Uni­ver­si­ty sci­en­tists and rese­ar­chers and to the state.”

Big Red 200 will be the first of Cray’s revo­lu­tio­na­ry new Shas­ta super­com­pu­ters instal­led at a U.S. uni­ver­si­ty. Lar­ger ver­si­ons of the­se sys­tems will be instal­led over the next few years at a num­ber of the federal Depart­ment of Energy’s labo­ra­to­ries as part of the Exas­ca­le Com­pu­ting Pro­ject, which aims to deve­lop the world’s fas­test super­com­pu­ters with Exas­ca­le speeds in excess of 1018 cal­cu­la­ti­ons per second.

The Big Red 200 sys­tem at India­na Uni­ver­si­ty will be an ama­zing resour­ce for advan­cing AI, sci­ence and engi­nee­ring, and will enab­le IU to play a pionee­ring role in the deve­lo­p­ment of app­li­ca­ti­ons, methods and soft­ware for Exas­ca­le com­pu­ting as a part­ner of Argon­ne Natio­nal Labo­ra­to­ry and DOE’s Exas­ca­le Com­pu­ting Pro­ject,” said Rick Ste­vens, one of the lea­ders of the ECP and the asso­cia­te labo­ra­to­ry direc­tor at Argon­ne Natio­nal Laboratory.

IU has got remar­kab­le value from its super­com­pu­ter sys­tems over the last 20 years,” said Brad Whee­ler, IU vice pre­si­dent for infor­ma­ti­on tech­no­lo­gy and chief infor­ma­ti­on offi­cer. “The­se sys­tems are hea­vi­ly used by hund­reds of IU rese­ar­chers, sci­en­tists and stu­dents in over 325 disci­pli­nes across the uni­ver­si­ty, and they have hel­ped fuel a rapid esca­la­ti­on of com­pe­ti­ti­ve rese­arch awards to IU.”

In 2018 alo­ne, IU was awar­ded more than $185 mil­li­on in rese­arch grant awards direct­ly sup­por­ted by the university’s high-per­for­mance com­pu­ting sys­tems. The $9.6 mil­li­on Big Red 200 sys­tem is fun­ded by reve­nue from federal con­tracts and grants, and it will be housed in IU’s sta­te-of-the-art Data Center.

We are exci­ted to con­ti­nue and expand our part­ners­hip with India­na Uni­ver­si­ty,” Cray CEO Peter Unga­ro said. “As the first uni­ver­si­ty to deploy Cray’s next-genera­ti­on Shas­ta super­com­pu­ting sys­tem, it will be intro­du­cing exas­ca­le-class capa­bi­li­ties to its rese­ar­chers across the con­ver­ging fiel­ds of mode­ling, simu­la­ti­on, ana­ly­tics and AI. Shas­ta will dri­ve exci­ting new inno­va­tions on cam­pus and in the sta­te of India­na and help enab­le the use of AI across their many diver­se rese­arch disci­pli­nes inclu­ding cyber­se­cu­ri­ty, medi­ci­ne and more.”

IU rese­ar­chers make exten­si­ve use of IU’s advan­ced com­pu­ting sys­tems and look for­ward to moving their rese­arch teams to Big Red 200.

Health and medi­cal are­as are one of the fas­test-gro­wing rese­arch domains for advan­ced machi­ne lear­ning app­li­ca­ti­ons. Big Red 200 will sup­port IU’s clo­se part­ners­hips bet­ween IU Health and rese­ar­chers at the IU School of Medi­ci­ne to dis­co­ver and advan­ce insights for qua­li­ty care. Com­pu­ta­tio­nal sys­tems for advan­ced AI rese­arch are beco­m­ing incre­a­singly cri­ti­cal for recrui­t­ing and retai­ning the best rese­arch talent in the state.

I am exci­ted about uti­li­zing the AI capa­bi­li­ties of Big Red 200 to acce­le­ra­te the rese­arch pro­grams in the Depart­ment of Medi­cal and Mole­cu­lar Gene­tics at the IU School of Medi­ci­ne,” said Dis­tin­guis­hed Pro­fes­sor Tatia­na Foroud, chair of the depart­ment. “Rese­ar­chers are now able to gene­ra­te lar­ge quan­ti­ties of mul­ti-omic data that is ide­al­ly sui­ted for AI app­li­ca­ti­ons. Ana­ly­ses powe­red by the expan­ded AI capa­bi­li­ties will allow sci­en­tists in the Depart­ment of Medi­cal and Mole­cu­lar Gene­tics to make novel dis­co­ve­ries elu­ci­da­ting the under­ly­ing mecha­nisms and pathways of dise­a­se. I belie­ve this new AI-capa­ble super­com­pu­ter will enab­le bre­akthrough dis­co­ve­ries across a broad ran­ge of rese­arch are­as, inclu­ding neu­ro­de­ge­ne­ra­ti­on and the stu­dy of Alzheimer’s disease.

Import­ant­ly, Big Red 200 will be an essen­ti­al resour­ce for the Pre­cisi­on Health Initia­ti­ve, one of the India­na Uni­ver­si­ty Grand Chal­len­ges, which is desi­gned to enhan­ce the pre­ven­ti­on, tre­at­ment and health out­co­mes of human dise­a­ses through a more pre­cise ana­ly­sis of the gene­tic, deve­lo­p­men­tal, beha­vio­ral and envi­ron­men­tal fac­tors that shape an individual’s health.”

David Crand­all, asso­cia­te pro­fes­sor of infor­ma­tics and com­pu­ting and direc­tor of the Com­pu­ter Visi­on Lab in the School of Infor­ma­tics, Com­pu­ting and Engi­nee­ring, added, “Big Red II came online in 2013 at the per­fect moment, just as the deep lear­ning revo­lu­ti­on was gai­ning steam and GPU com­pu­ting was beco­m­ing essen­ti­al for AI rese­arch. It allo­wed my lab to do rese­arch in com­pu­ter visi­on that would­n’t have other­wi­se been pos­si­ble. But cut­ting-edge AI models and algo­rith­ms now requi­re even grea­ter com­pu­ta­tio­nal power.

The new system’s latest-genera­ti­on V100 GPUs will allow my group — and many others throughout the uni­ver­si­ty — to con­ti­nue pushing for­ward the sta­te of the art in machi­ne lear­ning while app­ly­ing AI to important inter­di­sci­pli­na­ry app­li­ca­ti­ons. It will also be a fan­tastic tea­ching resour­ce for our cour­ses in machi­ne lear­ning, arti­fi­cial intel­li­gence, robo­tics and com­pu­ter vision.

A resour­ce of this sca­le is well bey­ond the bud­get of any one lab or group. I’m real­ly thril­led that the uni­ver­si­ty is con­ti­nuing its tra­di­ti­on of sup­por­ting rese­arch and edu­ca­ti­on through for­ward-loo­king invest­ments in cyberinfrastructure.”

Assi­stant pro­fes­sor Ben Kra­vitz, from the IU Bloo­m­ing­ton Col­le­ge of Arts and Sci­en­ces’ Depart­ment of Earth and Atmo­s­phe­ric Sci­en­ces, will also bene­fit from access to Big Red 200. He cur­r­ent­ly uses IU’s Big Red II super­com­pu­ter to model the effects of envi­ron­men­tal and cli­ma­te chan­ge throughout the world and in Indiana.

Kra­vitz uses glo­bal cli­ma­te model simu­la­ti­ons of various sce­n­a­ri­os of cli­ma­te chan­ge and is working on downs­ca­ling them into regio­nal models for India­na and the Mid­west. The­se data can then be used by other rese­ar­chers and prac­ti­tio­ners to under­stand how cli­ma­te chan­ge could affect water resour­ces, agri­cul­tu­re, emer­gen­cy plan­ning and even public health.

I’m a cli­ma­te mode­ler, and IU’s com­mit­ment to making advan­ced super­com­pu­ters avail­ab­le to facul­ty, staff and stu­dents is one of the rea­sons I took the job here,” Kra­vitz said. “I can do so much ama­zing work becau­se IU wants to make it happen.”

Fran­co Pestil­li, asso­cia­te pro­fes­sor of phy­sio­lo­gi­cal and brain sci­en­ces, works to acce­le­ra­te dis­co­very by con­nec­ting mas­si­ve brain data­sets and brain sci­ence rese­ar­chers across the world through brainlife.io.

Big Red 200 will beco­me an unpre­ce­den­ted uni­ver­si­ty asset to the endea­vors of brain sci­en­tists as it hel­ps enab­le fas­ter and bet­ter under­stan­ding of the human brain,” he said. “The­se efforts will likely advan­ce dia­gnostics and cure of a varie­ty of dise­a­ses, such as blind­ness (macu­lar dege­ne­ra­ti­on, glau­co­ma), brain aging and Alzheimer’s, and trau­ma­tic brain inju­ry and sports con­cus­sion, among others. This new excep­tio­nal resour­ce will bene­fit sci­ence and human society.”

IU’s Cen­ter for App­lied Cyber­se­cu­ri­ty Rese­arch will be one of the first groups to use Big Red 200.

With the app­li­ca­ti­on of AI to data-inten­si­ve cyber­se­cu­ri­ty, the­se rese­arch fron­tiers requi­re advan­ced sys­tems to sca­le our defen­ses to meet the high speed of today’s cyber thre­ats,” said Von Welch, IU exe­cu­ti­ve direc­tor for cyber­se­cu­ri­ty inno­va­ti­on. “This advan­ce­ment in IU’s cybe­r­in­fra­st­ruc­tu­re will allow us to advan­ce the sta­te of AI in cyber­se­cu­ri­ty in IU’s lea­ding-edge cyber­se­cu­ri­ty ser­vices: Omni­SOC and ResearchSOC.”

About India­na University
Foun­ded in 1820, India­na Uni­ver­si­ty is one of the world’s fore­mo­st public insti­tu­ti­ons. With more than 94,000 stu­dents and 21,000 employees state­wi­de, IU con­ti­nues to pur­sue its core mis­si­ons of edu­ca­ti­on and rese­arch while buil­ding a foun­da­ti­on for the university’s endu­ring strengths in tea­ching and lear­ning, world-class scho­l­ar­s­hip, inno­va­ti­on, crea­ti­ve acti­vi­ty, com­mu­ni­ty enga­ge­ment and aca­de­mic free­dom. Bloo­m­ing­ton is the flagship cam­pus of the uni­ver­si­ty, and each one of IU’s seven cam­pu­ses is an accredi­ted, four-year degree-gran­ting institution.

About Cray
Cray Inc. com­bi­nes com­pu­ta­ti­on and crea­ti­vi­ty so visio­na­ries can keep asking ques­ti­ons that chal­len­ge the limits of pos­si­bi­li­ty. Drawing on more than 45 years of expe­ri­ence, Cray deve­lo­ps the world’s most advan­ced super­com­pu­ters, pushing the bounda­ries of per­for­mance, effi­ci­en­cy and sca­la­bi­li­ty. Cray con­ti­nues to inno­va­te today at the con­ver­gence of data and dis­co­very, offe­ring a com­pre­hen­si­ve port­fo­lio of super­com­pu­ters, high-per­for­mance sto­rage, data ana­ly­tics and arti­fi­cial intel­li­gence solutions.

Durch die weitere Nutzung der Seite stimmst du der Verwendung von Cookies zu. Weitere Informationen

Die Cookie-Einstellungen auf dieser Website sind auf "Cookies zulassen" eingestellt, um das beste Surferlebnis zu ermöglichen. Wenn du diese Website ohne Änderung der Cookie-Einstellungen verwendest oder auf "Akzeptieren" klickst, erklärst du dich damit einverstanden.