{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,10]],"date-time":"2026-01-10T07:00:19Z","timestamp":1768028419844,"version":"3.49.0"},"reference-count":37,"publisher":"Wiley","issue":"6","license":[{"start":{"date-parts":[[2008,8,6]],"date-time":"2008-08-06T00:00:00Z","timestamp":1217980800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Concurrency and Computation"],"published-print":{"date-parts":[[2009,4,25]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>In the early 1990s, researchers at Sandia National Laboratories and the University of New Mexico began development of customized system software for massively parallel \u2018capability\u2019 computing platforms. These <jats:italic>lightweight kernels<\/jats:italic> have proven to be essential for delivering the full power of the underlying hardware to applications. This claim is underscored by the success of several supercomputers, including the Intel Paragon, Intel Accelerated Strategic Computing Initiative Red, and the Cray XT series of systems, each having established a new standard for high\u2010performance computing upon introduction. In this paper, we describe our approach to lightweight compute node kernel design and discuss the design principles that have guided several generations of implementation and deployment. A broad strategy of <jats:italic>operating system specialization<\/jats:italic> has led to a focus on <jats:italic>user\u2010level resource management<\/jats:italic>, <jats:italic>deterministic behavior<\/jats:italic>, and <jats:italic>scalable system services<\/jats:italic>. The relative importance of each of these areas has changed over the years in response to changes in applications and hardware and system architecture. We detail our approach and the associated principles, describe how our application of these principles has changed over time, and provide design and performance comparisons to contemporaneous supercomputing operating systems. Copyright \u00a9 2008 John Wiley &amp; Sons, Ltd.<\/jats:p>","DOI":"10.1002\/cpe.1361","type":"journal-article","created":{"date-parts":[[2008,8,6]],"date-time":"2008-08-06T11:26:42Z","timestamp":1218022002000},"page":"793-817","source":"Crossref","is-referenced-by-count":24,"title":["Designing and implementing lightweight kernels for capability computing"],"prefix":"10.1002","volume":"21","author":[{"given":"Rolf","family":"Riesen","sequence":"first","affiliation":[]},{"given":"Ron","family":"Brightwell","sequence":"additional","affiliation":[]},{"given":"Patrick G.","family":"Bridges","sequence":"additional","affiliation":[]},{"given":"Trammell","family":"Hudson","sequence":"additional","affiliation":[]},{"given":"Arthur B.","family":"Maccabe","sequence":"additional","affiliation":[]},{"given":"Patrick M.","family":"Widener","sequence":"additional","affiliation":[]},{"given":"Kurt","family":"Ferreira","sequence":"additional","affiliation":[]}],"member":"311","published-online":{"date-parts":[[2008,8,6]]},"reference":[{"key":"e_1_2_10_2_2","doi-asserted-by":"publisher","DOI":"10.1145\/146941.146944"},{"key":"e_1_2_10_3_2","doi-asserted-by":"crossref","first-page":"275","DOI":"10.1155\/1994\/602764","article-title":"PUMA: An operating system for massively parallel systems","volume":"3","author":"Wheat SR","year":"1994","journal-title":"Scientific Programming"},{"key":"e_1_2_10_4_2","volume-title":"Proceedings of the 1995 Intel Supercomputer User's Group Conference","author":"Shuler L","year":"1995"},{"issue":"2","key":"e_1_2_10_5_2","first-page":"4","article-title":"Dynamic processor modes in PUMA","volume":"8","author":"Maccabe AB","year":"1996","journal-title":"Bulletin of the Technical Committee on Operating Systems and Application Environments (TCOS)"},{"key":"e_1_2_10_6_2","unstructured":"NagH GotfriedR GreenbergDS KimC MaccabeAB Mack StallcupT LaddG ShulerL WheatSR van DresserD.PROSE: Parallel real\u2010time operating system for secure environments. Proceedings of the Intel Supercomputer Users Group Conference Knoxville TN June 1996."},{"key":"e_1_2_10_7_2","doi-asserted-by":"crossref","unstructured":"PetriniF KerbysonD PakinS.The case of the missing supercomputer performance: Achieving optimal performance on the 8 192 processors of ASCI Q. Proceedings of SC'03 Phoenix AZ 2003.","DOI":"10.1145\/1048935.1050204"},{"key":"e_1_2_10_8_2","doi-asserted-by":"publisher","DOI":"10.1007\/3-540-45825-5_50"},{"key":"e_1_2_10_9_2","unstructured":"BrightwellR HudsonT RiesenR MaccabeAB.The Portals 3.0 message passing interface. Technical Report SAND99\u20102959 Sandia National Laboratories December1999."},{"key":"e_1_2_10_10_2","volume-title":"Workshop on Communication Architecture for Clusters CAC'02","author":"Brightwell R","year":"2002"},{"key":"e_1_2_10_11_2","unstructured":"BrightwellR LevenhagenM MaccabeAB RiesenR.A performance comparison of Myrinet protocol stacks. Proceedings of Third Linux Clusters Institute Conference on Linux Clusters St. Petersburg FL 2002."},{"key":"e_1_2_10_12_2","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2006.65"},{"key":"e_1_2_10_13_2","unstructured":"RiesenR BrightwellR MaccabeAB HudsonT PedrettiK.The Portals 3.3 message passing interface: Document revision 2.0. Technical Report SAND2006\u20100420 Sandia National Laboratories January2006."},{"key":"e_1_2_10_14_2","doi-asserted-by":"publisher","DOI":"10.1145\/602770.602868"},{"key":"e_1_2_10_15_2","doi-asserted-by":"crossref","unstructured":"BrightwellR RiesenR UnderwoodK BridgesPG MaccabeAB HudsonT.A performance comparison of Linux and a lightweight kernel. Proceedings of the 2003 IEEE International Conference on Cluster Computing (Cluster 2003) Hong Kong China December 2003.","DOI":"10.1109\/CLUSTR.2003.1253322"},{"key":"e_1_2_10_16_2","unstructured":"WallaceD.Compute Node Linux: Overview progress to date and roadmap. Proceedings of the 2007 Cray User Group Annual Technical Conference May 2007."},{"key":"e_1_2_10_17_2","doi-asserted-by":"publisher","DOI":"10.1007\/3-540-57981-8_168"},{"key":"e_1_2_10_18_2","unstructured":"BaileyD BarszczE BartonJ BrowningD CarterR DagumL FatoohiR FinebergS FredericksonP LasinskiT SchreiberR SimonH VenkatakrishnanV WeeratungaS.The NAS parallel benchmarks. Technical Report RNR\u201094\u2010007 NASA Ames Research Center Moffett Field CA 1994."},{"key":"e_1_2_10_19_2","unstructured":"CookJ EnbodyR HerlandBG.SUNMOS on the Intel Paragon: Evaluation and early experience. Technical Report pl\u2010jc\u201094.0 Department of Informatics University of Bergen March1994."},{"key":"e_1_2_10_20_2","unstructured":"GreenbergD MaccabeB McCurleyKS RiesenR WheatS.Communication on the Paragon. Proceedings of the Intel Supercomputer Users' Group 1993. Annual North America Users' Conference 1993;117\u2013124."},{"key":"e_1_2_10_21_2","unstructured":"HertelESJr BellRL ElrickMG FarnsworthAV KerleyGI McGlaunJM PetneySV SillingSA TaylorPA YarringtonL.CTH: A software family for multi\u2010dimensional shock physics analysis. Proceedings of the 19th International Symposium on Shock Waves Marseille France July 1993;377\u2013382."},{"key":"e_1_2_10_22_2","doi-asserted-by":"crossref","unstructured":"TsafrirD EtsionY FeitelsonDG KirkpatrickS.System noise OS clock ticks and fine\u2010grained parallel applications. ACM International Conference on Supercomputing Cambridge MA June 2005.","DOI":"10.1145\/1088149.1088190"},{"key":"e_1_2_10_23_2","unstructured":"CampWJ TomkinsJL.Thor's hammer: The first version of the Red Storm MPP architecture. Proceedings of the SC 2002 Conference on High Performance Networking and Computing Baltimore MD November 2002."},{"key":"e_1_2_10_24_2","doi-asserted-by":"publisher","DOI":"10.1145\/582034.582071"},{"key":"e_1_2_10_25_2","doi-asserted-by":"publisher","DOI":"10.1177\/1094342005056114"},{"key":"e_1_2_10_26_2","doi-asserted-by":"crossref","unstructured":"LampsonBW.Hints on computer system design. Proceedings of the 9th ACM Symposium on Operating Systems Principles Bretton Woods New Hampshire 1983;33\u201348.","DOI":"10.1145\/773379.806614"},{"key":"e_1_2_10_27_2","doi-asserted-by":"publisher","DOI":"10.1145\/1341312.1341316"},{"key":"e_1_2_10_28_2","volume-title":"Cray Research MPP Software Guide","author":"Cray Research, Inc.","year":"1994"},{"key":"e_1_2_10_29_2","unstructured":"AuslanderM FrankeH KriegerO GamsaB StummM.Customization\u2010lite. Proceedings of the 6th Workshop on Hot Topics in Operating Systems (HotOS\u2010VI ) Cape Cod MA 1997;43\u201348."},{"key":"e_1_2_10_30_2","unstructured":"http:\/\/www.research.ibm.com\/K42[23 May2008]."},{"key":"e_1_2_10_31_2","doi-asserted-by":"crossref","unstructured":"ChenD.QCDSP: A teraflop scale massively parallel supercomputer. Supercomputing'97 San Jose CA November1997.","DOI":"10.1145\/509593.509645"},{"key":"e_1_2_10_32_2","unstructured":"http:\/\/www.eecg.toronto.edu\/parallel\/tornado.html[23 May2008]."},{"key":"e_1_2_10_33_2","doi-asserted-by":"crossref","unstructured":"MoreiraJ BrutmanM CastanosJ GoodingT InglettT LieberD McCarthyP MundyM ParkerJ WallenfeltB GiampapaM EngelsiepenT HaskinR.Designing a highly\u2010scalable operating system: The Blue Gene\/L story. Proceedings of the 2006 ACM\/IEEE International Conference for High\u2010Performance Computing Networking Storage and Analysis (SC'06) Tampa FL November 2006.","DOI":"10.1109\/SC.2006.23"},{"key":"e_1_2_10_34_2","doi-asserted-by":"crossref","unstructured":"OldfieldRA MaccabeAB ArunagiriS KordenbrockT RiesenR WardL WidenerP.Lightweight I\/O for scientific applications. Proceedings 2006 IEEE Conference on Cluster Computing Barcelona Spain September 2006.","DOI":"10.1109\/CLUSTR.2006.311853"},{"key":"e_1_2_10_35_2","doi-asserted-by":"publisher","DOI":"10.1145\/1131322.1131330"},{"key":"e_1_2_10_36_2","doi-asserted-by":"crossref","unstructured":"BrunetonE CoupayeT LeclercM QuemaV StefaniJB.An open component model and its support in Java. Seventh International Symposium CBSE Edinburgh U.K. 2004.","DOI":"10.1007\/978-3-540-24774-6_3"},{"key":"e_1_2_10_37_2","unstructured":"FassinoJF StefaniJB LawallJ MullerG.Think: A software framework for component\u2010based operating system kernels. USENIX 2002 Annual Conference Monterey CA 2002."},{"key":"e_1_2_10_38_2","volume-title":"Model Driven Engineering for Distributed Real\u2010time Embedded Systems","author":"Tournier JC","year":"2005"}],"container-title":["Concurrency and Computation: Practice and Experience"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.wiley.com\/onlinelibrary\/tdm\/v1\/articles\/10.1002%2Fcpe.1361","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/cpe.1361","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,10,10]],"date-time":"2023-10-10T23:12:01Z","timestamp":1696979521000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/cpe.1361"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2008,8,6]]},"references-count":37,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2009,4,25]]}},"alternative-id":["10.1002\/cpe.1361"],"URL":"https:\/\/doi.org\/10.1002\/cpe.1361","archive":["Portico"],"relation":{},"ISSN":["1532-0626","1532-0634"],"issn-type":[{"value":"1532-0626","type":"print"},{"value":"1532-0634","type":"electronic"}],"subject":[],"published":{"date-parts":[[2008,8,6]]}}}