g-Social: Enhancing Integrated e-Science Tools with Social Networking Functionality

During the last decade, the scientific community has witnessed an unprecedented deployment of large-scale, federated e-Infrastructures such as Grid Computing, primarily for supporting data-intensive scientific exploration and coordinated problem solving. However, practical experience and user studies have indicated that the adoption of such e-Infrastructures is lagging behind original expectations, a fact which is mainly attributed to the limited support that available tools provide for user collaboration and information sharing. The goal of this paper is twofold, first to lay down the foundations for building a collaboration environment in the form of abstractions and second to show the effectiveness of these abstractions through g-Social, an Eclipse-based, open-source environment as an extension to g-Eclipse, that provides a powerful, user-friendly, platform-independent toolset for users, application developers and administrators of Grid infrastructures. g-Social enables user collaboration and resource sharing through Online Social Networking services, capitalizing on the success that these services have.

Collaboration with agents in VR environments

Jan Ciger

Virtual reality is gaining on importance in many fields – scientific simulation, training, therapy and also more and more in entertainment. All these applications require the human user to interact with virtual worlds inhabited by intelligent characters and to solve simulated or real problems. This thesis will present an integrated approach to simulated problem solving in virtual reality environments, with the emphasis on teamwork and the ability to control the simulations. A simulation framework satisfying these goals will be presented. A unified approach to the representation of semantic information in virtual environments based on predicate calculus will be introduced, including the representation of the world state, action semantics and basic axioms holding in the simulated world. Afterwards, the focus will be on the collaboration model based on task delegation and facilitator-centric architecture. A simple but efficient facilitator design will be presented. The issues of the collaborative problem solving will be examined. A new technique using propositional (STRIPS-like) planning with delegated actions and object-specific planning will be described. A control technique for virtual characters/objects will be detailed, enabling run-time exchange of control and control sharing over a virtual entity between multiple autonomous agents and/or human users. Finally, a set of case studies will be shown, illustrating the possible applications of the techniques developed and described in this dissertation.

EPFL2005

Operating System and Network Co-Design for Latency-Critical Datacenter Applications

Evangelos Marios Kogias

Datacenters are the heart of our digital lives. Online applications, such as social-networking and e-commerce, run inside datacenters under strict Service Level Objectives for their tail latency. Tight latency SLOs are necessary for such services to remain interactive and keep users engaged. At the same time, datacenters operate under a single administrative domain which enables the deployment of customized network and hardware solutions based on specific application requirements. Customization enables the design of datacenter-tailored and SLO-aware mechanisms that are more efficient and have better performance. In this thesis we focus on three main datacenter challenges. First, latency-critical, in-memory datacenter applications have ÎŒs-scale services times and run on top of hardware infrastructure which is also capable of ÎŒs-scale inter-node round-trip times. Existing operating systems, though, were designed under completely different assumptions and are not ready for ÎŒs-scale computing. Second, the base of datacenter communications is Remote Procedure Calls (RPCs) that depend on a message-oriented paradigm, while TCP still remains widely-used for intra datacenter communications. The mismatch between TCPâs bytestream-oriented abstraction and RPCs causes several inefficiencies and deteriorates tail latency. Finally, datacenter applications follow a scale-out paradigm based on large fan-out communication schemes. In such a scenario, tail latency becomes a critical metric due to the tail at scale problem. The two main factors that affect tail latency is interference and scheduling/load balancing decisions. To deal with the above challenges we advocate for a co-design of network and operating system mechanisms targeting ÎŒs-scale tail optimisations for latency-critical datacenter applications. Our approach investigates the potential of pushing functionality to the network leveraging emerging in-network programmability features. Whenever existing abstractions fail to meet the ÎŒs-scale requirements or restrict our design space, we propose new ones given the design and deployment freedom the datacenter offers. This thesis contributions can be split in three main parts. We, first, design and build tools and methodologies for ÎŒs-scale latency measurements and system evaluation. Our approach depends on a robust theoretical background in statistics and queueing theory. We, then, revisit existing operating system and networking mechanisms for TCP-based datacenter applications. We design an OS scheduler for ÎŒs-scale tasks, while we modify TCP to improve L4 load balancing, and provide an SLO-aware flow control mechanism. Finally, after identifying the problems related to TCP-based RPC services, we introduce a new transport protocol for datacenter RPCs and in-network policy enforcement that enables us to push functionality to the network. We showcase how the new protocol improves the performance and simplifies the implementation of in-network RPC load balancing, SLO-aware RPC flow control, and application-agnostic fault-tolerant RPCs.

EPFL2020

Operating System and Network Co-Design for Latency-Critical Datacenter Applications

Evangelos Marios Kogias

EPFL2020