Organizational Culture

From The Business of Energy in U.S. Academic Research Computing
Jump to: navigation, search

Introduction to Organizational Culture

Culture exists any time humans gather together, whether in businesses, at school, or at home. Culture provides a set of tools and filters by which we experience one another. We are simultaneously part of many cultures, and sometimes we refer to subsets of a dominant culture as sub-cultures. For example, a manager at an academic research computing center must operate effectively as a member of the management hierarchy’s culture of her university, the workplace culture of the center, the peer culture of other directors, and the broader national or international culture of similar centers.

She is also likely to be part of the culture of supercomputing users, such as the faculty, staff and students that engage in scientific discovery using supercomputers. At any moment, she, like all humans, will adapt to the different signals, settings, and situations she encounters to help understand what is going on, to effectively communicate, and to achieve goals. Meanwhile, the broader societal cultures (i.e., western society, national, state, regional and local cultures) play a role, as well as non-workplace cultures (i.e., hobbies, religion, family, and so forth).

Organizational culture is a topic of deep interest in the business community. Business scholars seek to understand how organizational culture can be assessed and, potentially, modified. Bremer (2012), for example, devotes much of her book to describing an organizational culture assessment instrument. The instrument, which rests largely on survey tools, is intended to measure organizational culture, and to assess where intervention might benefit an organization.

Colleges and universities that make up higher education in the US have been studied extensively for organizational culture and other characteristics, and books have been written about the diversity, nuances and challenges of such institutions. In the social sectors addendum to Jim Collins' Good to Great (2005), he observes wryly that “a faculty is 10,000 points of ‘no.’” This comment gets to the heart of many universities, where there is a culture of freedom, honed by a passion for debate.

Factors such as budget pressures and uncertainty about measuring the return on investment for supercomputing, as discussed elsewhere, can result in cultural pressures for research computing centers. In the following sections, we seek to provide some guidance on how to navigate the changing environment.

CASC as a Culture Keeper for Research Computing

CASC is a major forum for communication among leaders of research computing units, drawing dominantly from academic supercomputing centers or similar organizations in the United States. CASC members get frequent email updates on topics of potential interest, and assemble twice per year for a program focusing on the industry, the commercial and government sectors we interact with most, and on aspects of daily concern such as disk storage policies and the latest software.

Equally important is that CASC serves to inform members that they are not alone. Most states have just a few institutions with research computing facilities (some have one, or none), and there is a vast range in the number of people, size and types of systems, and user constituency at each center. CASC serves to bring the industry together under what binds us, including passion for computationally based scientific discovery, devotion to outstanding user service, and interests in leading-edge technologies.

As with most cultures, the CASC role for keeping the culture of research computing is often not explicitly stated. As observed by Edgar Schein (2009) in his book on corporate culture, I believe CASC is forever at the center of a roiling intersection of cultures. Some member sites are in early startup phase, seeking the university equivalent of venture capital and angel investors. Others are highly mature, with a more stable role within their institutions. Turnover among personnel at all levels is relatively high, as experts find their skills are portable and, sometimes, under-appreciated at home.

Stewardship as a Key Goal of Organizational Culture in Research Computing

The notion of “stewardship” for research computing was highlighted by Berente & Rubleske’s (2013) analysis of a survey of the CASC membership. Stewardship is described on page 19:

“Note that in place of leadership we used the construct of a ‘culture of stewardship.’ This culture is fostered from the organizational leaders, but spans multiple leaders and is more a function of the organization’s history than any particular leader’s personality. Our interview‐based research into academic computing organizations identified stewardship - the psychological ownership / championing of the academic computing domain - as important to academic computing organizations.

Our findings, which are detailed in the following pages, the culture of stewardship is the only consistent predictor [of their outcome variable].”

The survey focused on different notions of innovation as the major outcome variable for their regression analysis. For the purposes of this discussion of culture, I find the emphasis on culture of stewardship to be the most important message for leaders of research computing centers to take to heart. Stewardship involves all the characteristics that might be found in a center director’s job description: effectiveness at managing personnel and budgets, highly functioning systems for users and other stakeholders, caretaking and periodic technology refresh for the equipment, and high quality outcomes such as scientific papers and student success.

Berente & Rubleske’s notion of stewardship goes an important additional step, identifying the need for stewardship to transcend the center and its leader, and extend to the broader organization: the many leaders and other stakeholders that make up the institution. Even beyond the institution, to stakeholders that might include state legislatures, regents, trustees, major donors, regional businesses, the media, and others. The paper observes a statistically strong relationship between stewardship as they measured it, and innovations. As the survey is refined, I have little doubt that lack of stewardship will be identified as an indicator of current or pending hardship.

Challenges of Maintaining Stewardship

Research computing is expensive, requiring costly equipment, specialized facilities to house them, and highly trained staff. Despite all the advances in computing, supercomputers are still rather challenging to utilize effectively. Successful campus- or regional-serving centers might have a few hundred users, working on several scores of projects. As discussed elsewhere, assessing return on investment is an ongoing challenge.

Due to the dynamic and varied nature of cultures at institutions of higher education, there is a constant need to assess the multiple levels of culture that impact the center. Are there disgruntled users? Have students been successful in their work? To what extent are deans and other division leaders aware of the benefits the center provides?

Stewardship extends from the center up through the institution’s ranks. Reporting might be to a dean, a vice chancellor or provost, or directly to the chancellor or other top executive. To what extent is there visible and sustained commitment to stewardship of research computing by those important figures in the hierarchy?

Competing interests abound in university settings. In times of plenty, research computing might be seen as an easy way to invest capital in highly visible instrumentation. In lean times, supercomputers are up against everything from electron microscopes to classroom technologies when budget cuts are being considered. As the ROI analysis elsewhere demonstrates, a multifaceted, institutionally-aligned, and ongoing approach to identifying the value of research computing is needed.

Even then, it might not be enough. Is communication of scientific outcomes at an appropriate level for the stakeholders? Are there vocal proponents? We have often seen that many supercomputer users, like geeks everywhere, might not be great spokespeople - nor are their best students. In cultures of high competition, research computing might be facing off against theater, business, or any of a multitude of endeavors that might have broader appeal. Or, it might be against other capital needs such as fixing buildings or repairing roads.

ROI analysis can be helpful to shape the desired understanding of the benefits of research computing. Stewardship, as a manifestation of organizational culture, also has important emotional and interpersonal aspects. These, too, need to be cultivated.

Lest this seem daunting, especially to a new, struggling center: there is a smaller-scale way. In his writings on the topic, Gifford Pinchot defines an “intrapreneur” as a change agent within an existing, often large, organization (e.g., Pinchot & Pellman, 1999). Fundamentally, research computing leaders in all but the most well-established centers (and probably most of them, too) operate as intrapreneurs within their organizations.

Such leaders seek constantly to position their activities relative to shared goals of the institution for scientific, academic and social outcomes. Fundamentally, research computing centers provide a commons for discovery and analysis. The technology lifecycle, along with constant innovation in science, software, and analysis, leads to a constant reinvention.

Pinchot recommends that any intrapreneur must have a “sponsor,” which is someone who will repeatedly make a case for the value of what the change agent is doing. The sponsor must occupy a higher stratum in the organization, but need not be directly in the leader’s chain of command. Such sponsors can be invaluable for protecting centers, and their leaders, from competition.

Staffing for Research Computing

There is overlap between staff qualifications for research computing and other computer-based organizations (i.e., campus computing), and the academic disciplines that make use of the center's facilities.

The smallest centers might only have a few people, devoting part of their time to keeping the systems up and running. Centers with larger staff sizes will often separate user-facing services (i.e., helpdesk and various levels of consulting) from inward-facing services (i.e., systems administration and similar duties). Larger centers are more likely to include some computational scientists, who can work deeply with users to assist them in getting their computational work done.

Employee Development

To recruit and obtain the high quality staff members needed for a well-functioning research computing organization, I strongly encourage center directors to engage directly in employee development.

Most universities have rather poor prospects for employee advancement, and sometimes those prospects are especially bleak for technology-focused employees. There are some straightforward steps that centers can take to provide better prospects for their staff:

  • Work with the university human relations office to pursue available mechanisms for recognition and advancement.
  • Promote from within when possible.
  • Cultivate student workers. They might become your best employees, perhaps after they have gone elsewhere to gain additional experience.
  • Never stop training: encourage constant career development in all employees. Even if there is no budget for training, staff members can read a book or watch a video or attend a seminar, and report back to each other.
  • Be a good neighbor. An unfortunate aspect of the lack of a career path at universities means that changing jobs might be the best way for an employee to advance their career. Work closely with your employees, and potential recruits, to provide a gentle transition from place to place. Better still, provide an adequate compensation scheme and career path to the people you have.
  • Give appreciation and reward to the staff, and celebrate the contributions that each staff member makes.

References

Berente, Nicholas & Rubleske, Joseph. 2013. “Academic computing management benchmarks: Beta version.” Online: [1]. Accessed February 2, 2014.

Bremer, Marcella. 2012. Organizational Culture Change. Zeolle, Netherlands: Kikker Groep.

Collins, Jim. 2005. Good to Great and the Social Sectors: A Monograph to Accomany Good to Great. New York: HarperCollins.

Pinchot, Gifford & Pellman, Ronald. 1999. Intrapreneuring in Action: A Handbook for Business Innovation. San Francisco: Berrett-Koehler.

Schein, Edgar. 2009. The Corporate Culture Survival Guide. Hoboken: Jossey-Bass.