[Rahul Ramachandran]

[toc]

 

 

RDA Big Data Interest Group Charter

Mission

The ultimate goals of RDA Big Data Interest Group is to produce a set of recommendation documents to advise diverse research communities with respect to:

How to select an appropriate Big Data solution for a particular science application with optimal value? and

What are the best practices in dealing with various data and computing issues associated with such a solution?

Objectives

In order to achieve our mission, we need to attain the following objectives, while take duly into account related activities and results, such as those of International Organization for Standardization (ISO), Open Geospatial Consortium (OGC), and US National Institute of Standards and Technology (NIST) Big Data Public Working Group (NBD-PWG), as well as other relevant organizations and undertakings.

• Clarifying, and sometimes defining, terminologies related to Big Data.

  • Any Big Data solution for scientific research will involve many relevant disciplines such as computing hardware/software infrastructure and architecture, data management/curation, analyses and algorithm, etc. Discussions will be more effective when there is no confusion in terminology.

  • The efforts of the RDA Terminologies WG, as well as other relevant efforts (e.g. from NDB-PWG), will be consulted and feedback will be provided whenever necessary.

• Characterizing leading Big Data technologies.

  • Investigations will be carried out 1) directly through spin-off Working Groups (WGs) and 2) in collaboration with other RDA groups, to characterize the technologies.

  • The characterization of a Big Data technology will include its strengths, weaknesses, and limitations. In other words, what it is good for, in what sort of environment, for what kind of analyses/algorithms.

  • Example evaluation criteria include, but not limited to,

    • Performance, resource utilization, and scalability,

    • Usability,

    • Flexibility and extensibility, and

    • Propensity in supporting scientific collaborations.

• Collaborating with external entities through IG member involvements.

  • These external organizations and enterprises include, but not limited to, ISO, OGC, NIST, EarthCube, EarthServer, or even individual research projects.

  • These interactions with external entities will ensure BDIG is up to date with new developments and enable us to leverage others’ efforts.

  • Examples of such interactions include: the establishment of connection with NIST NBD-PWG activities, participation in its discussion of Big Data reference architecture, as well as the exercise of common Big Data use cases collected by NBD-PWG and BDIG; similar activities with OGC on Big Geo Data and the ISO SQL WG on flexible retrieval from Big Science Data.

• Producing a set of recommendation documents based on results obtained from activities in attaining above objectives.

  • This set of documents will include:

    • A systematic classification of algorithms pertinent to the characterization of Big Data technologies,

    • Characterizations of Big Data technologies investigated, especially their value characteristics in each category of use cases,

    • Frequency of each class of algorithms and/or queries used by workflows in various use cases, delineated by science domains/subdomains, and

    • Feasible combinations of analysis algorithms, analytical tools, data and resource characteristics and scientific queries.

  • These recommendation documents are aimed to serve as a best practice guide for scientific groups/communities interested in investing in Big Data technologies.

Participation

Big Data Interest Group, BDIG,  is open to all RDA members to participate. The following participants are especially relevant:

• Domain scientists wishing to utilize Big Data solutions for their research and/or applications,

• Data specialists with experience in data production, curation, analysis, and management, especially involving large volumes and varieties of data,

• Computational scientists or software engineers with special interests in data analysis techniques and algorithm analysis, especially pertaining to Big-Data-relevant technologies and tools,

• Experts, or aspiring experts, of various Big Data technologies and tools,

• Computational infrastructure and architecture experts in fields such as distributed computing, high-performance computing, and database systems,

• Data scientists with a blended interest involving some subsets the activities mentioned above, in partcular with share, use, and reuse of open scientific data collections, and

• Managers involved in any combination of the activities mentioned above.

Interaction Mechanism

BDIG will utilize capabilities provided by the RDA platform to communicate and collaborate effectively to achieve its goals. These include:

• Monthly telecoms/webex to with planned agenda to discuss specific issues

• Asynchronous collaboration using google docs, wiki and email list servers

• Semiannual RDA Plenary meetings to hold sessions for F2F interactions amongst members and to inform other RDA members of its ongoing activities.

Outcome

BDIG will be considered a success if the Interest Group:

• Develops recommendation documents accepted within and beyond RDA

• Visibility and uptake of results can be demonstrated within RDA, but also beyond RDA (such as OGC and ISO, discussions and outreach of results at specific domain-specific events like the IGARSS, EGU, AGU, etc.)

All BDIG documents will be publicly available on the RDA server.

Schedules

[The following is a tentative schedule of planned activities. It will be amended, updated, and published when group members have reviewed and agreed upon it.]

Year	Quarter	Task
2015	Q1	Revise BDA IG (original group name) charter to suit the broadened scope of proposed IG name change to “Big Data Interest Group”. Prepare for RDA 5th Plenary.
	Q2	Start the planning and organization of studies into the characterization of various popular Big Data technologies. Evaluation of potential WG spin-offs based on the characterization plan.
	Q3	Progress reports on characterization studies. Prepare RDA 6th Plenary. Establish WGs for detailed Big Data studies.
	Q4	Progress reports on characterization studies.
2016	Q1	Produce a report on characterization studies. Prepare RDA 7th Plenary. Initial reports of results from spin-off WGs and their findings.
	Q2

 

References

[1] NIST Big Data Use Cases: http://bigdatawg.nist.gov/usecases.php

[2] EarthServer Big Data Use Cases: http://www.earthserver.eu/Services

[3] OGC Big Data Domain Working Group: http://external.opengeospatial.org/twiki_public/BigDataDwg/

Recommendation document [3/14 – 3rd RDA Plenary]

Discussion