Share |

Why aren’t there more women in HPC?

Working towards gender equality in the supercomputing community

The Women in HPC network supports collaboration and networking by bringing together female HPC scientists, researchers, developers, users, and technicians from across the UK.

In late 2013, several members of staff at the Edinburgh Parallel Computing Centre (EPCC) in the UK realized that despite many initiatives in industry, academia, and government to improve the representation of women in science, technology, engineering, and mathematics, things did not seem to be improving in the field of high-performance computing (HPC). As a discipline that spans many traditional subjects, we had expected gender representation to change as schemes such as Athena SWAN and Project Juno and groups such as BCS Women and IoP Women in Physics started to have an impact. However, our personal experience suggested that things did not appear to be changing.

Not a day goes by when equality, women’s rights, equal pay or the ‘glass ceiling’ aren’t in the news. So why set up yet another group addressing gender equality when there are plenty of initiatives already? Our impression was that, so far at least, none of these initiatives were having an impact in our field.

Bringing women together and improving visibility

Women in HPC supports collaboration and networking by bringing together female HPC scientists, researchers, developers, users, and technicians. The initiative also encourages women working in HPC to engage in outreach activities and improve the visibility of inspirational role models. A major part of our work is to understand the global gender demographics of the HPC community. We need to identify the barriers to women engaging with HPC, establish why women choose to leave the field, and share best practices between countries, industry, and academia.

Possibly the most important part of our work is bringing women together. Providing women with the opportunity to meet female role models, potential mentors, and peers who are experiencing the same sorts of problems as themselves is a crucial part of our work. Men in a male-dominated environment will naturally have male peers that they can talk to and discuss issues with; for women in the same environment that is much harder. Our events aim to bring women together, to allow them to connect and continue a relationship online to gain the support that they might not otherwise receive.

Toni Collis is the coordinator of the Women in HPC initiative. Image courtesy the University of Edinburgh.

Golden opportunities to improve — and the importance of doing so

My background in physics means I am used to being surrounded by male peers. At school I was always more comfortable around boys than girls, and although I was very much aware that I was unusual — a girl studying physics and later taking an interest in computing — I didn’t feel uncomfortable. What really opened my eyes was realizing that in the majority of meetings I attend as an HPC scientist I am the only woman in the room. Occasionally in such meetings it is assumed that I am there in an administrative capacity rather than because I work in HPC. I also realized that I am more likely to participate in discussions when there are other women in the room and I am not alone in feeling this way.

As a discipline that spans multiple scientific fields, HPC has a unique opportunity to address gender inequality. HPC users include biologists, chemists, engineers, physicists, climate scientists, and researchers from many more disciplines. In 2010/11, 60% of biology postgraduate degrees in the UK were awarded to women; the corresponding figures for physics and computer sciences are 41% and 20%, respectively. The gender demographics of the user communities could therefore help to improve diversity within the field of HPC.

Sadly, initial analysis is not promising: only 17% of the UK national HPC service users are women, 9% of EASC 2013 and 5% of PGAS 2013 attendees were female, and only 11% of attendees at the largest 2014 HPC conference in the world, SC14, were women (based on those attendees who responded to the gender question in the SC14 registration questionnaire). Anecdotal evidence also supports this: many of the women I speak to are often the only woman in their group.

A complex issue

The reasons that women are less likely to pursue science, academia, and research are multifaceted, are difficult to isolate, and therefore are difficult to quantify individually. Why women are less likely to pursue HPC is a just a different aspect of the problem that is being studied extensively around the world, but understanding it will help us improve diversity for our community.


If you would like to find out more about the Women in HPC initiative, including information about upcoming events and how to become a member, please visit our website.

Your rating: None Average: 3.4 (7 votes)


Post new comment

By submitting this form, you accept the Mollom privacy policy.