Open source is increasingly important to many organizations'
smart technology strategies. Two new offerings from IBM Research and the Milepost
consortium aim to make it even smarter.
The Milepost GCC
open-source machine learning compiler lets users develop, test and
optimize applications up to 10 times faster than current tools,
according to partners. (Note: IBM sponsors Smarter Technology but does
not control editorial content.) The first-of-its-kind
offering, hailed as “the world’s first intelligent, open-source compiler” debuted
June 25.
“Our technology automatically learns how to get the
best performance from the hardware—whether mobile phones, desktops or entire
systems, the software will run faster and use less energy," Bilha
Mendelson, manager of code optimization technologies at IBM Research Haifa,
said in the release. "We opened the compiler environment so it can access
artificial intelligence and machine learning guidance to automatically determine
exactly what specific optimizations should be used and when to apply them to
ramp up performance."
Also introduced: a companion “Collective
Tuning Center”—a collaborative
community portal dedicated to creating software more quickly. Users share optimization
best practices, organizer said, and development of common R&D tools with
open APIs aimed at automating program optimization, compiler design and
architecture tuning using empirical, statistical and machine learning
techniques.
The project consortium
includes the IBM Haifa Research Lab, Israel; the University of Edinburgh, United Kingdom;
ARC International, United Kingdom; CAPS Enterprise, France; and INRIA, France.
The offerings are noteworthy
for several reasons. They can potentially:
Speed SW roll out and enterprise innovation. The universal mandate to “do more
with less” has expanded to “faster too.” As the developers note: In many organizations, software developers are
fast becoming the nucleus of innovation, crucial to all business processes.
They build the services and capabilities that will underlie future revenue and
generate business opportunity.
Ease strains on IT resources. The average enterprise devotes
30 percent to 50 percent of its entire technology infrastructure to the development and
testing of software, the partners said. Anything that reduces that burden frees
precious resources for higher-value work.
Advance intelligent software development. What we’re talking about here is
the shift from specialized compilers to automated code optimization. Unlike commercial
compilers, The Milepost GCC uses machine learning to
automatically optimize programs for reconfigurable heterogeneous embedded
processors. Leveraging AI is a potentially important and
long overdue way to combat the twin challenges of increasing development time
and power usage and decreasing performance. Milepost could spark renewed
interest in the long-overdue practical
application of AI in code development.
Shorter software development times, bigger performance
gains. What’s not to like? Intelligent
compilers may not be the sexiest new smart technology. But history shows that big
technology advances often are hidden under the hood. Hopefully Milepost will
prove the first marker on a long new road to better, faster, greener software
development.
