40th Anniversary of COMPSAC: Highlights

Sorel Reisman, California State University

Pages: 4–6

Abstract—The 40th IEEE Computer Society International Conference on Computers, Software & Applications was held in 2016 in Atlanta, Georgia, from June 10–14. The author reviews this year's conference theme, keynote addresses, and significant panels and how they relate to IT professionals.

Keywords—COMPSAC; software engineering; big data; Internet of Things

The Computers, Software, and Applications Conference (COMPSAC), a Signature Conference of the IEEE Computer Society (CS), celebrated its 40th anniversary this year in Atlanta, Georgia, from June 10–14. This year also marks the 70th anniversary of the CS.

As a signature CS conference, sponsored directly by the CS Board of Governors, COMPSAC's mission is to serve the professional interests of all our members, both young and old, those new to the profession and also well-established researchers and practitioners. This year's conference theme, “Connected World: New Challenges for Data, Systems, and Applications,” covered the entire breadth of CS members’ fields of interest—from architecture to software engineering, including application development through to social networking.

COMPSAC 2016 was well received. It attracted 300 computer and IT academics and professionals from more than 25 countries.

In addition to more than 200 double-blind peer-reviewed presentations and workshops on several emerging topics whose papers will be soon available in IEEE's digital library, Xplore (http://ieeexplore.ieee.org), this year's conference featured two new programs: The first was a full day Special IEEE Fellows Panel led by COMPSAC co-general chair and CS past president Dejan Milojicic—himself an IEEE Fellow. The second was the IEEE Young Professionals Special Panel entitled “The IoT and the Connected World We Live In,” organized and moderated by Ramesh Nair and Prasanth Mohan, both members of the CS Membership and Geographic Activities (MGA) board.


The goal of the Fellows Panel was, as Milojicic asserts, to “mine the untapped creative potential of IEEE's richest intellectual assets—IEEE Fellows” to help develop strategic directions for various IEEE and CS initiatives and activities. This year, the panel focused on COMPSAC itself, and how the conference can continue to serve our diverse membership in a rapidly changing, technology-driven world. Among its recommendations was to conduct a CS “Rock Stars” event at next year's conference. This will be consistent with the 2017 conference theme of “Building Digital Autonomy for a Sustainable World.”

Next year, the conference will be held in Torino, Italy, July 4–8, and will highlight, in addition to other dimensions of autonomy, diverse technical and nontechnical aspects of autonomous vehicles. The 2017 venue, Politecnico di Torino, is the ideal site for this because the institution serves as a center of research for many automobile companies, including General Motors, now the owner of the Italian automobile brand Fiat.

At the other end of the age demographic from Fellows was the Young Professionals panel. Its focus was on the Internet of Things (IoT), providing a spotlight on the broadening computer platform that will one day encompass planet Earth—technologies they and their successors will inherit and via which they hope to capitalize on various autonomously moving modes of transportation.


Probably among the most noteworthy events of the conference were the keynote addresses from three of the world's top experts in computer architecture and communications technologies. Their presence and interactions, both with each other and with attendees, proved both lively and fascinating.

Multilayer Computing Model

The first keynoter was Yale N. Patt, the internationally renowned computer architecture pioneer and distinguished professor of electrical and computer engineering at the University of Texas at Austin, and a longtime friend of the CS. His address, “If Moore's Law Does in Fact End, Whose Job Is It to Pick Up the Slack? And How?,” in addition to talking about the end of Moore's law, also humorously but seriously described, among other things, a hierarchical, multilayer model of “computing,” ranging from a top layer of “problem” definition down to the layer of “electrons.” His model maps most topics inherent in a freshman computer science curriculum, including algorithms, programming, instruction set architecture, microarchitecture, and circuits, and an integrated and nonsequential approach to teaching and learning these key topics—which would produce a more holistically educated computer science student.

In Patt's discussion of Moore's law, he challenged attendees, mostly software people this year, to address software solutions to extend the life of aging processing technologies.

Memory-Driven Computing

Kirk Bresniker, chief architect of systems research at Hewlett Packard Labs, where he guides research and advanced development of novel hardware and software system designs, presented the second keynote. Bresniker, who is an HP Enterprise Fellow, spoke about “Memory Driven Computing—Adapting Algorithms to Thrive on Abundant Memory.” He told the audience that Moore's law might be ready to end just when the IoT will demand more processing than ever before. He also took the audience through the history of chip design as it relates to Moore's law, demonstrating that while circuits have indeed shrunk incredibly since Gordon E. Moore, the co-founder of Intel and Fairchild Semiconductor, first postulated the law, in fact, most of the shrunken logic went to increasing memory size rather than to substantially increasing processor speed. And physically shrunken but expanded memory accounts for most of the perceived doubling of processor speed every 12–24 months.

He went on to describe HP's The Machine which is “a new kind of computer that allows you to do things you can't even conceive today” (www.labs.hpe.com/research/themachine/). As outlined on its webpage,

The Machine puts the data first. Instead of processors, we put memory at the core of what we call “Memory-Driven Computing.” Memory-Driven Computing collapses the memory and storage into one vast pool of memory called universal memory. To connect the memory and processing power, we're using advanced photonic fabric. Using light instead of electricity is key to rapidly accessing any part of the massive memory pool while using much less energy.

Ethernet Memory Lane

Our third keynoter, Bob Metcalfe, the inventor of Ethernet, spoke on “Ethervation: Lessons from 43 Years of Ethernet Innovation.” His presentation was a terrific walk down memory lane for many of us as he chronicled the history of Ethernet, and his role in conceiving its fundamentals, productizing it, and eventually monetizing the technology.

At dinner the evening before his talk, Metcalfe facetiously commented to Bresniker that photonic access to memory might preclude the use of technologies such as his (now the world's) Ethernet. But of course, he was kidding, since for the foreseeable future, photonic access isn't in the cards for “remote” devices such as the computer I’m using right now, or the trillions of IoT devices that really will need that kind of processor access.

These days, Metcalfe is Professor of Innovation at Cockrell School of Engineering, University of Texas at Austin—the same university and the same department as Yale Patt!—where Metcalfe says his mission is “to help Austin and now Baustin be better Silicon Valleys.”

IT in Practice Symposium

One of the new additions to COMPSAC last year was a “symposium”—essentially a souped-up conference track, labeled ITiP—Information Technology in Practice, led by San Murugesan, the editor in chief of IT Professional magazine. The ITiP symposium this year featured three sessions comprising two invited talks and peer-reviewed presentations.

Simon Y. Liu, past editor in chief of IT Pro and current associate administrator for the Agricultural Research Service at the US Department of Agriculture, talked about “Data-Intensive Research and Scientific Discovery.” He outlined a new paradigm of scientific discovery based on data-intensive research, and illustrated how it can be realized through a few real-world research projects at the Agricultural Research Service.

A longtime IT Pro advisory board member, a frequent submitter to IT Pro, and now the state of Washington ICT Industry Sector Policy Lead, Joseph Williams gave an interesting talk entitled “How IT Leadership Will Change by 2020 and Why It Matters.” Looking ahead to 2020, he cautioned that deep shifts in the technology landscape will evolve the role of the IT leader in ways that would have been unimaginable just 10 years ago. These deep shifts are going to change the careers of many IT professionals. In fact, this year, the editorial and advisory boards of IT Pro met at the conference and reviewed the magazine's history and current status and discussed articles and themes to address these kinds of changes.

Limited space here, plus the impossibility of my being at all the conference sessions, prevents me from providing more detail about other sessions, speakers, and events—however, all of them are chronicled in the archives of the conference webpage at www.compsac.org.

So, if you weren't one of the attendees at this incredible conference, you really missed an intellectually stimulating and fun event. Never mind! I invite you all to actively participate in the conference next year. Visit the COMPSAC 2017 website, note all the scheduled call-for-paper dates, and present your recent work and insights. And we'll see you in Torino. Ciao, Baby!

Sorel Reisman is COMPSAC Standing Committee Chair and IT Pro Advisory Board Chair. He is also managing director of MERLOT.org, president emeritus of the IEEE Computer Society, professor at California State University, and a Fulbright OER Specialist. Contact him at sreisman@computer.org.
66 ms
(Ver 3.x)