The participants in these work- hoo! have made large-scale clusters
shops are primarily researchers. The available to the academic community
workshopsalsoinvolverepresentatives for education and research, and the
of funding agencies—critical to transi- National Science Foundation has an-tioning research visions into funded nounced its CluE (Cluster Exploratory)
programs. Often they also involve in- research initiative. There is no magic
dustrial participants. A recent example here—it takes dedicated individuals to
of success is CCC’s “Big Data Comput- make things happen. But CCC can be
ing Study Group.” In late March 2008, an enabler.
two workshops were held in Sunnyvale, A number of other CCC activities are
CA. The first was the “Hadoop Sum- described on CCC’s Web site, which
mit,” whose goal was to build a com- includes descriptions of various grand
munity of users of Hadoop, an open- challengeproblemsandablogdevoted
source version of Google’s MapReduce to discussions of research visions for
system1 for distributing computations the field. More broadly, CCC is work-across clusters of thousands of nodes. ing along with other organizations to
The second was the “Data-Intensive “get the word out” regarding our field.
Scalable Computing Symposium,” I encourage you to become engaged.
whose goal was to build a community Participate in the CCC research visions
of researchers concerned with various blog ( www.cccblog.org/). Join with col-issues related to “big-data computing” leagues to propose a workshop to chart
(slides, videos, and summaries are a compelling vision for future of your
linked from the CCC Web site; www. subfield.
cra.org/ccc/). Both of these communi- Computer science has accom-ty-building exercises were successful. plished so much, and there is so much
And, as a result of preliminary work additional exciting work to do. The op-
TOdIotn_ehablyft_hpeagceo_re4Cg_rmoaurpkso.fpodrfg:aLnaiyzeorust 1por6t/u2n6i/ti0e8s ar3e:t2ru8lyPeMxtra Poar gdein1ary. It’s
of this effort, Google, IBM, and Ya- up to us to seize these opportunities.
References
1. Dean, J. and Ghemawat, S. MapReduce: Simplified
data processing on large clusters. In Proceedings of
the Sixth Symposium on Operating System Design and
Implementation (OSDI ’04), (San Francisco, CA, Dec.
2004); labs.google.com/papers/mapreduce.html.
2. Evolving the High Performance Computing and
Communications Initiative to Support the nation’s
Information Infrastructure. Computer Science
and Telecommunications Board, National Research
Council, 1995; www7.national-academies.org/cstb/
pub_hpcci.html.
3. Grand Challenges for Engineering. National Academy
of Engineering; www.engineeringchallenges.org.
4. Innovation in Information Technology. Computer
Science and Telecommunications Board, National
Research Council, 2003; www7.national-academies.
org/cstb/ pub_itinnovation.html.
5. The Computing Community Consortium: Who,
what, when, where, why, and how. Computing
Research news 20, 1 (Jan. 2008); www.cra.org/CRN/
issues/0801.pdf.
Ed Lazowska ( lazowska@cs.washington.edu) is the Bill &
Melinda Gates Chair in Computer Science & Engineering
at the University of Washington and the chair of the
Computing Community Consortium.
© 2008 ACM 0001-0782/08/0800 $5.00
ACM Transactions on
Internet Technology
�����
This quarterly publication encompasses many disciplines in
computing—including computer software engineering, middleware, database management, security, knowledge discovery
and data mining, networking and distributed systems,
communications, and performance and scalability—all under
one roof. TOIT brings a sharper focus on the results and roles
of the individual disciplines and the relationship among
them. Extensive multi-disciplinary coverage is placed on the
new application technologies, social issues, and public policies
shaping Internet development.
�����
http://toit.acm.org/