Dear Colleague,
Many thanks for supporting the International Webinar Series. As we are
following GMT, first, please check the time zones in your country and
make sure you follow the timings correctly. Please read the following
instructions carefully for the smooth operations of the daily
technical program. Please join Webex using one of the methods below.
Meeting link: https://mir.webex.com/mir/j.php?MTID=m34eeb3471f71c939ce9f09fd8b3696e1
Meeting number: 166 175 6636
Password: pdJpwYNh322
Join by video system
Dial [log in to unmask]
You can also dial 210.4.202.4 and enter your meeting number.
Join by phone
+65-6703-6949 Singapore Toll
Access code: 166 175 6636
1. Please make sure that you have the adequate browser / plugins etc.
to run Webex. It is better to use your laptop / desktop computer and
mute your mobile phones during the webinar.
2. Please assemble not earlier than 10 minutes and try to assemble on
time. On July 31, we have the following speaker:
Dr. Leszek Sliwko, UK Home Office, United Kingdom (July 31, 14:00 15:00 GMT)
TITLE: AI-driven Load Balancer for Cluster Systems'
BIOGRAPHY: Leszek is a highly qualified IT professional (with 16
years’ experience in Scala/Java development and Cloud systems
architecture) and has a solid understanding of software development
life cycle. He has worked in multiple sectors namely: the UK
Government, Investment Banking, Financial Consultancy and Telecoms. He
currently leads a team of Scala programmers and testers in developing
and maintaining several internal UK Government systems. Leszek is an
excellent communicator and his assignments involve stakeholder
management. He has worked on projects where he was responsible for
regular correspondence and coordination with high-profile external
partners such as the United Nations High Commissioner for Refugees,
the International Organization for Migration, and the UK Cabinet
Office. Leszek is keen on experimenting with the latest technologies,
especially those related to High-Performance Computing and Machine
Learning. In 2019 he was awarded a Ph.D. in Parallel and Distributed
Computing for his work developing a novel AI-driven load balancer for
Cloud systems.
ABSTRACT: Cloud computing is an established technology allowing users
to share resources on a large scale. A cloud system connects multiple
individual servers in order to process related tasks in several
environments at the same time. The sheer size of the system itself
means that thousands of machines may be involved. The focus of this
research was to design a strategy to dynamically allocate tasks
without overloading Cloud nodes which would result in system stability
being maintained at minimum cost. During this exploration, a
high-fidelity Cloud workload simulator, based on a month-long workload
traces from Google's computing cells, was created. Two possible
approaches to resource management were proposed and examined: the
centralised metaheuristic load balancer and the decentralised
AI-driven system. The project involved extensive experiments run on
the HPC cluster, and the promising results are presented together with
detailed discussions and a conclusion.
3. Attendee interaction during the session: As we are dealing with
nearly 1000 participants, it is important that all attendees mute the
microphone soon after joining the session. Please don't flood the
chat windows with greetings and thank you messages. There will be a Q
and A session soon after every talk. Attendees may place their
questions (only very important and that might be also useful for the
whole audience) in the chat window and the moderator will select only
few of them (depending on the time availability).
4. We will distribute participation certificates for each session
during the following day. Participation certificates will be shared on
a drive. Each attendee will have to complete an online form within 12
hours after the completion of the session. Please take adequate time
to provide your personal details / information as further correction
requests will not be entertained. Please note that the feedback form
will be given in the Webex chat window soon after every session. So it
is mandatory to stay until the end of the session and also complete
all sections in the feedback form to receive the certificate. You may
discard filling the feedback form, if you don't need the participation
certification.
Our resources are limited and we also have to go for four weeks. So we
kindly request your understanding and support as we might not be able
to serve any individual queries related to the webinars.
Thank you all and we look forward to seeing you during the Webinars.
Please stay safe and healthy!
Yours sincerely,
Prof. (Dr.) Ajith Abraham
Director - Machine Intelligence Research Labs (MIR Labs)
Scientific Network for Innovation and Research Excellence
11, 3rd Street NW
P.O. Box 2259
Auburn, Washington 98071, USA
http://www.mirlabs.org
Email: [log in to unmask]
Tel: +1(253) 2453001 / +1(646) 8016894
Personal WWW: http://www.softcomputing.net
ORCID: https://orcid.org/0000-0002-0169-6738
SCOPUS: https://www.scopus.com/authid/detail.uri?authorId=7202760099
Publons: https://publons.com/researcher/2897631/ajith-abraham/
G-Scholar: http://scholar.google.com/citations?user=i95DGLQAAAAJ
Researchgate: https://www.researchgate.net/profile/Ajith_Abraham2
DBLP: https://dblp.org/pers/hd/a/Abraham:Ajith
########################################################################
To unsubscribe from the SUPPORT-VECTOR-MACHINES list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=SUPPORT-VECTOR-MACHINES&A=1
This message was issued to members of www.jiscmail.ac.uk/SUPPORT-VECTOR-MACHINES, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
|