Call for papers: Dynamic Distributed Data-Intensive Applications, Programming Abstractions, and Systems

02.12.10 | Comment?

We’re look­ing for papers on the top­ics of pro­gram­ming with large, dynamic data for a work­shop co-located with HPDC in San Jose next year.
Call for papers

Work­shop on Dynamic Dis­trib­uted Data-Intensive Applic­a­tions, Pro­gram­ming Abstrac­tions, and Sys­tems (3DAPAS)

To be held in con­junc­tion with HPDC-2011, 8 June 2011, San Jose, CA


There has been a lot of effort in man­aging and dis­trib­ut­ing tasks where com­pu­ta­tional loads are dom­in­ant. Such applic­a­tions have after all, been his­tor­ic­ally the drivers of “grid” com­put­ing.  There has, how­ever, been rel­at­ively less effort on tasks where the com­pu­ta­tional load is matched by the data load, or even dom­in­ated by the data load. For such tasks to be able to oper­ate at scale, there are con­cep­tu­ally simple run-time trade-offs that need to be made, such as determ­in­ing whether to move data to com­pute versus keep­ing data loc­al­ized and move com­pu­ta­tional tasks to oper­ate on the data in situ, or pos­sibly neither, and with data regen­er­ated on-the-fly. Due to fluc­tu­at­ing resource avail­ab­il­ity and cap­ab­il­it­ies, as well as insuf­fi­cient prior inform­a­tion about applic­a­tion require­ments, such decisions must be made at run-time. Fur­ther­more, resource, con­nectiv­ity and/or stor­age con­straints may require the data to be manip­u­lated in-transit so that it is “made-right” for the con­sumer. Cur­rently it is very dif­fi­cult to imple­ment these dynamic decisions or the under­ly­ing mech­an­isms in a general-purpose and scal­able fashion.

Although the increas­ing volumes and com­plex­ity of data will make many prob­lems data load dom­in­ated, the com­pu­ta­tional require­ments will still be high.  In prac­tice, data-intensive applic­a­tions will encom­pass data-driven applic­a­tions.  For example, many data-driven applic­a­tions will involve com­pu­ta­tional activ­it­ies triggered as a con­sequence of inde­pend­ently cre­ated data; thus it is imper­at­ive for an applic­a­tion to be able to respond to unplanned changes in data load or con­tent.  There­fore, under­stand­ing how to sup­port dynamic com­pu­ta­tions is a fun­da­mental, but cur­rently miss­ing ele­ment in data-intensive computing.This work­shop will oper­ate at the triple point of dynamic and dis­trib­uted and data-intensive (3D) attrib­utes. This work­shop will oper­ate at the triple point of dynamic, dis­trib­uted and data-intensive (3D) attrib­utes. It will also focus on innov­at­ive approaches for scalab­il­ity in the end-to-end real-time pro­cessing of sci­entific data. We refer to 3D applic­a­tions as those are data-intensive, need to sup­port and respond to dynamic data, and, either are fun­da­ment­ally, or need to be, dis­trib­uted. We are inter­ested in papers that span the spec­trum from the design of cyber­in­fra­struc­ture to sup­port 3D applic­a­tions, to novel applic­a­tion examples. We are also look­ing to bring research­ers together to look at hol­istic, rather than piece­wise, approaches to the end-to-end pro­cessing and man­aging of sci­entific data.

3DAPAS builds upon a 3 year research theme on Dis­trib­uted Pro­gram­ming Abstrac­tions (DPA), which has held a series of related work­shops (see: DPA Past Events) includ­ing but not lim­ited to e-Science2008, Euro­Par 2008 and the CLADE series. 3DAPAS will also draw on ideas from the ongo­ing 3DPAS Research Theme fun­ded by the NSF and UK EPSRC.

Top­ics of interest include but are not lim­ited to:

  • Case stud­ies of devel­op­ment, deploy­ment and exe­cu­tion of rep­res­ent­at­ive 3D applications
  • Pro­gram­ming sys­tems, abstrac­tions, and mod­els for 3D applications
  • What are the com­mon, min­im­ally com­plete, char­ac­ter­ist­ics of 3D application?
  • What are major bar­ri­ers to the devel­op­ment, deploy­ment, and exe­cu­tion of 3D applic­a­tions? What are the primary chal­lenges of 3D applic­a­tions at scale?
  • What pat­terns exist within 3D applic­a­tions, and are there com­mon­al­it­ies in the way such pat­terns are used?
  • How can pro­gram­ming mod­els, abstrac­tion and sys­tems for data-intensive applic­a­tions be exten­ded to sup­port dynamic data applications?
  • Tools, envir­on­ments and pro­gram­ming sup­port that exist to enable emer­ging dis­trib­uted infra­struc­ture to sup­port the require­ments of dynamic applic­a­tions (includ­ing but not lim­ited to stream­ing data and in-transit data analysis)
  • Data-intensive dynamic work­flow and in-transit data manipulation
  • Abstrac­tions and mech­an­isms for dynamic code deploy­ment and “mov­ing the code to the data”
  • Applic­a­tion drivers for end-to-end sci­entific data management
  • Runtime sup­port for in-situ analysis
  • Sys­tem sup­port for high end workflows
  • Hybrid com­put­ing solu­tions for in-situ analysis
  • Tech­no­lo­gies to enable multi-platform workflows
Sub­mis­sion Require­ments:
Authors are invited to sub­mit tech­nical papers of at most 8 pages in PDF format, includ­ing all fig­ures and ref­er­ences. Papers should be format­ted in the ACM Pro­ceed­ings Style and sub­mit­ted via Easy­Chair. Accep­ted papers will appear in the con­fer­ence pro­ceed­ings, and will be incor­por­ated into the ACM Digital Library.

Sub­mis­sion of a paper implies that at least one author will attend the work­shop to present the paper, if it is accepted.

Papers must be self-contained and provide the tech­nical sub­stance required for the pro­gram com­mit­tee to eval­u­ate the paper’s con­tri­bu­tion. Papers should thought­fully address all related work. Sub­mit­ted papers must be ori­ginal work that has not appeared in and is not under con­sid­er­a­tion for another con­fer­ence or a journal. See the ACM Prior Pub­lic­a­tion Policy for more details.

Import­ant Dates:

Sub­mis­sions Due: 31 Jan 2011
Paper Decisions Announced: 28 Feb 2011
Final Camera-Ready Papers Due: 24 Mar 2011

Work­shop Date: 8 June 2011

(all dates are firm)


  • Daniel S. Katz, Uni­ver­sity of Chicago & Argonne National Labor­at­ory, USA
  • Shantenu Jha, Louisi­ana State Uni­ver­sity, USA & e-Science Insti­tute, UK
  • Jon Weiss­man, Uni­ver­sity of Min­nesota, USA

Pro­gramme Com­mit­tee Mem­bers:

  • Gab­ri­elle Allen, Louisi­ana State Uni­ver­sity, USA
  • Mal­colm Atkin­son, eSI & Uni­ver­sity of Edin­burgh, UK
  • Henri Bal, Vrije Uni­versiteit, Netherlands
  • Jon Blower, Read­ing e-Science Centre, Uni­ver­sity of Read­ing, UK
  • Shawn Brown, Uni­ver­sity of Pitt­s­burgh & Pitt­s­burgh Super­com­put­ing Cen­ter, USA
  • Simon Dob­son, Uni­ver­sity of St. Andrews, UK
  • Den­nis Gan­non, Microsoft, USA
  • Keith R. Jack­son, Lawrence Berke­ley National Lab, USA
  • John R. John­son, Pacific North­w­est National Labor­at­ory, USA
  • Scott Klasky, Uni­ver­sity of Ten­nessee & Oak Ridge National Labor­at­ory, USA
  • Ber­tram Ludäscher, Uni­ver­sity of Cali­for­nia, Davis, USA
  • Abani Patra, Uni­ver­sity of Buf­falo, USA
  • Man­ish Para­shar, Rut­gers & NSFUSA
  • Omer Rana, Cardiff Uni­ver­sity, UK
  • Joel Saltz, Emory Uni­ver­sity, USA
  • Domen­ico Talia, Uni­versita’ della Calab­ria, Italy

Related Posts: