high-performance computing
caring commitments. Others suggest that women are more likely to be there because they are viewed as better at networking and representing their business. What is clear is that, at this point, no one knows. What we do know is that attending conferences is good for an individual’s career. Conferences facilitate the building of networks, keep people up to date with the most recent research and findings, and offer the opportunity to discuss and present your work, thereby improving a persons standing in the community.
Tus it is important that everyone has an equal chance to attend conferences and events if they wish to. If there are barriers to participation in place for any group, irrespective of gender or any other personal characteristic, these need to be identified and removed or minimised. EuroMPI has never before collected gender of attendees or considered the gender of authors. Tis is a similar approach to the majority of other conferences, and something that Women in HPC (WHPC) is attempting to change.
IT SHOULD BE NOTED
THAT THE RELATIONSHIP BETWEEN THE
EuroMPI 2016 conference to act as diversity chair. EuroMPI, now in its 23rd year, is the pre-eminent conference for the current dominant method of programming parallel architectures – message passing and the associated API – the Message Passing Interface (MPI). If MPI is to rise not only to the challenge of helping HPC take the leap to exascale, but also to address the challenge of rapidly becoming a programming paradigm available to everyday programming thanks to the growing availability of small parallel clusters, MPI needs to adapt and change. Tis is a theme that EuroMPI 2016 will embrace as it addresses ‘the modern challenges to MPI’s dominance in HPC’. However, to embrace this theme and move forward effectively, the organisers recognise that women have to be involved in the debate.
It should be noted that the relationship between the proportion of women at a conference and the proportion of women in a community are not well understood. Are women equally, more or less likely to attend conferences than their male peers? Some suggest that women are less likely to attend conferences because of childcare/
www.scientific-computing.com l
PROPORTION OF WOMEN AT A CONFERENCE AND THE PROPORTION OF WOMEN IN A COMMUNITY ARE NOT WELL UNDERSTOOD
It has oſten proved controversial; collecting personal information that isn’t necessary should be avoided. Collecting any personal information can impact, on an individual causing effects such as stereotype threat to emerge, or be misconstrued to be used as a way to adhere to quotas. However, if we are to understand the representation of women in the HPC and MPI communities we need to know how many women there. So this year, for the first time, EuroMPI will ask registrants for their gender and report our findings in aggregate form aſter the conference. Tis can then be used in future years as a starting point to see if changes need to be made and if, when changes are implemented, they have any effect on the diversity of the attendees.
As it is the first time that EuroMPI is openly considering how to improve gender, our intention is not to hinder the
@scwmagazine
attendance of men, but rather to remove as many barriers as is realistically possible that may apply to women – many of which apply to men too, of course).
Te first thing we realised is that this is not an easy task! If it were an easy problem to solve, someone would already have done so; enough people spend time considering how to improve diversity that if there were a ‘silver bullet’ we would all have implemented it by now! Gender initiatives are oſten viewed with trepidation as no one wants to damage the integrity of the conference. Tere is oſten concern that recording gender of authors will be used to influence the outcome of the review process. We did record the gender of author submissions so we can identify the corresponding acceptance rate, but we set no goals; this is simply a recording process to help understand if women are currently well represented and how this changes in future years.
We also challenged ourselves to consider new and exciting speakers who have not been considered before. Again, we did not set quotas: this was fundamental to our approach, but we did need to look outside our immediate list of contacts. Tis has been a rewarding experience for the committee as it has encouraged us to meet new people and find out what they are doing. It has also resulted in a list of diverse and fascinating topics for the keynote programme.
Finally, the conference has become the first ever event to partner with Women in HPC. Te partnership requires the conference to take out the steps outlined above and in exchange, the conference is advertised to WHPC members and a WHPC event will run alongside (although this is open to all attendees of the conference, not just women). l
About EuroMPI
www.eurompi2016.ed.ac.uk
Now in its 23rd year, EuroMPI is the leading conference for users and developers of MPI. The aim of this conference is to bring together all of the stakeholders involved in developments and applications related to the Message Passing Interface (MPI). As the preeminent meeting for users, developers and researchers to interact and discuss new developments and applications of message-passing parallel computing, the annual meeting has a long, rich tradition. This year’s conference will be held in Edinburgh, Scotland from 25th to 28th September 2016.
About WHPC
www.womeninhpc.org
WHPC was created with the vision to educate and collaborate with the HPC community and encourage participation by providing knowledge, fellowship, and support to women and the organisations that employ the. Toni Collis is the Director and founder of the Women in HPC (WHPC) network in the UK.
References [1]
J. B. Bear and A. W. Woolley, ‘The Role of Gender in Team Collaboration and Performance’, Interdisciplinary Science Reviews, Vol. 36 No. 2, June, 2011, 146–53
AUGUST/SEPTEMBER 2016 13
ra2studio/
Shutterstock.com
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36
