Information Technology Reference
In-Depth Information
9.3 WORKFLOWSMEETING BIG DATA
There is a soaring quantity of information, also known as the data
deluge, in business, government, and science. In business, Wal-mart
transaction databases are estimated to contain more than 2.5 petabytes
of data [229], growing at a fast pace with customer behavior and
preferences, and market trends data. In the military, US Air Force
drones collected around 24 years' worth of video footage over Afgha-
nistan and Iraq in 2009, and the amount was supposed to be 30 times as
many in 2011 [229]. In science, in 2010 the detectors at the Large
Hadron Collider (LHC) facility at CERN produced 13 petabytes of data
[75]. Moreover, sensor, social media, mobile, and location data are
growing at an unprecedented speed. In parallel to their fast growth, data
are also becoming increasingly semi-structured and diversified.
This astonishing growth and diversity in data have profoundly
affected the way people process and make sense of them. How the
workflow technology should evolve for accessing, assembling, ana-
lyzing, and acting upon big data remains a big challenge. At build-
time, we need to investigate what needs to be changed in a workflow
meta-model, to describe a workflow that gathers, organizes, and
processes a large volume of data (structured, semi-structured, and
unstructured). Data are going to be the first-class citizen in workflows.
At run-time, a workflow engine needs to interact with the underlying
computation infrastructure (cloud or on-premise clusters) and data
processing mechanisms such as MapReduce [230]. The collaborative
interplay among the three, that is, workflow, data, and computation
infrastructure, is vital to enable the low-latency and high-throughput
analytics on big data.
The novel applications of SOA to help build business, scientific,
and medical workflows promise higher-level service reuse and better
process management for research organizations and business enter-
prises. The area requires much work to be performed by not only
academic researchers but also industrial developers and practitioners.
This topic has presented some of the recent research and development
results that shall help move the application of SOA a step forward. It
calls for more efforts into novel service composition methods and their
applications in business, scientific, and medical services and software
systems.
Search WWH ::




Custom Search