Merge pull request #265 from gliargovas/master

June 2024 talk and profile updates

content/members/gliargovas.md

@@ -7,20 +7,11 @@ id: m_gliargovas
 givenname: Georgios
 surname: Liargkovas
 email: [email protected]
-web_site: https://gliargovas.github.io
+web_site: https://liargkovas.com
+web_log: http://liargkovas.com/blog/
 github: gliargovas
 linkedin: george-liargovas-796259175
 twitter: gliargko
 photo: gliargovas.jpg
 
-**Georgios Liargkovas** graduated with a Bachelor's Degree from the Department of Management Science & Technology at the Athens University of Economics and Business (AUEB). Currently, he delves deep into research spanning a variety of areas.
-
-His prime research interests encompass:
-
-- **Computer Systems**, with a keen focus on operating system design and implementation, distributed systems, scheduling, concurrency, file systems, and kernel-level programming.
-
-- **Empirical Software Engineering**, particularly in the realms of software analytics and mining software repositories.
-
-- **Programming Languages**, where he has a growing interest in compilers and program analysis.
-
-- An emerging interest in **Machine Learning**, especially considering its potential synergy with the domains mentioned above.
+**Georgios Liargkovas** will be pursuing a PhD in OS scheduling and cloud computing, with a focus on distributed and serverless architectures, at Columbia University, advised by [Prof. Kostis Kaffes](https://www.cs.columbia.edu/~kkaffes/index.html). He graduated with a Bachelor's degree in Software Engineering and Data Science from the Department of Management Science and Technology at Athens University of Economics and Business. Since 2020, he has been a research assistant at [BALab](https://www.balab.aueb.gr/), advised by [Prof. Diomidis Spinellis](https://www.spinellis.gr/index.html), concentrating on empirical software engineering and mining software repositories studies. He is also an affiliate researcher at Brown University’s [Atlas Systems Group](https://atlas-group.cs.brown.edu/), engaged in advancing shell-script parallelization, advised by [Prof. Nikos Vasilakis](https://nikos.vasilak.is/). His research interests include system design and optimization, particularly through the application of machine learning techniques. Outside academia, he is passionate about long-distance running, cycling, and music.

content/seminars/2024-06-12.md

@@ -0,0 +1,13 @@
+title: Reordering Shell Script Execution for Enhanced Performance
+presenter: Georgios Liargkovas, AUEB and Brown University
+date: 2024-06-12
+time: 18:30
+category: seminars
+
+The execution order of shell scripts is determined by their syntax, thus missing key optimization opportunities routinely available in single-language runtime environments.
+hS is a new system that exploits the potential for non-linear execution in shell scripts, executing components speculatively out of order by inferring—and complying with—their runtime dependencies during their execution.
+To achieve this, hS introduces a new system-call monitor that collects ordering and effect constraints, a lightweight container that controls the order and application of side effects, a formally verified streaming scheduler that executes components within a window of speculation while respecting their ordering constraints, and several runtime optimizations for speculation and application of side effects.
+Applying hS to a large and diverse set of shell scripts yields a ~2x speedup for free (i.e. without any extra code or annotations).
+
+## Biography
+Georgios Liargkovas will be pursuing a PhD in OS scheduling and cloud computing, with a focus on distributed and serverless architectures, at Columbia University, advised by Prof. Kostis Kaffes. He graduated with a Bachelor's degree in Software Engineering and Data Science from the Department of Management Science and Technology at Athens University of Economics and Business. Since 2020, he has been a research assistant at BALab, concentrating on empirical software engineering and mining software repositories studies. He is also an affiliate researcher at Brown University’s Atlas Systems Group, engaged in advancing shell-script parallelization. His research interests include system design and optimization, particularly through the application of machine learning techniques. Outside academia, he is passionate about long-distance running, cycling, and music.