VU Computer Science Student Association - STORM
Wherever we turn, our society is digital. Online gaming and streaming, science and engineering, business-critical operations, and analytics depend, often transparently, on the inter-operation of many distributed computer systems. However successful, we cannot take these _ecosystems_ for granted: the core does not rely on sound principles of science and design, and there are warning signs about the scalability, dependability, and sustainability of operations. For applications like online gaming, cheating and other toxic activity can be rampant. It is important to keep this complex computing infrastructure inclusive, from the young digital explorer to the digital expert.
In this talk, inspired by these challenges and by our experience with distributed computer systems for over 15 years, we envision Massivizing Computer Systems, a domain of computer science focusing on understanding, deploying, scaling, and evolving successfully such ecosystems; all with fundamental challenges. The fundamental challenges also include the processes and tools to engineer ecosystems, aiming to enable computing and networking as a true utility.
We posit that we can address these fundamental challenges by focusing on computer ecosystems rather than merely on (individual, small-scale) computer systems. We can achieve this through an ambitious, comprehensive research program; by advancing and at times rethinking core concepts in distributed (eco)systems, such as observability, scalability, consistency, availability, and sustainability; by adapting and interlinking operational techniques from replication to data partitioning to autoscaling; by considering decision-making about infrastructure over both short-term (milliseconds to days) and long-term (months to years); yet aiming for operational simplicity wherever possible. This vision aligns with the Manifesto on Computer Systems and Networking Research in the Netherlands that the speaker co-leads.
Through examples related primarily to online gaming, we show how our approach works in practice, focusing on the interplay between the science, design, and engineering of massivizing computer systems. We cover four main pillars of modern gaming ecosystems, the virtual world, the analytics, the (procedural) content generation, and the meta-gaming (social) network. Many of our examples come from real-world prototyping and experimentation, grand experiments in computer systems, and/or benchmarking and performance analysis work conducted with the Cloud industry R&D group of SPEC.