Distributed Systems

The processing of large amounts of data on distributed execution platforms such as MapReduce or Heron contributes significantly to the energy consumption of today's data centers. The E³ project aims at minimizing the power consumption of such execution environments without sacrificing performance. For this purpose, the project develops means to make execution environments and data-processing platforms energy aware and to enable them to exploit knowledge about applications to dynamically adapt …

→ More information

Internet-based services play a central role in today's society. With such services progressively taking over from traditional infrastructures, their complexity steadily increases. On the downside, this leads to more and more faults occurring. As improving software-engineering techniques alone will not do the job, systems have to be prepared to tolerate faults and intrusions.

REFIT investigates how systems can provide fault and intrusion tolerance in a resource-efficient manner. The key technology…

→ More information

Coordination services such as ZooKeeper are essential building blocks of today's data-center infrastructures as they provide processes of distributed applications with means to exchange messages, to perform leader election, to detect machine or process crashes, or to reliably store configuration data. Providing an anchor of trust for their client applications, coordination services have to meet strong requirements regarding stability and performance. Only this way, it is possible to ensure that…

→ More information

Distributed Ledger Technologies (DLTs), often referred to as blockchains, enable the realisation of reliable and attack-resilient services without a central infrastructure. However, the widely used proof-of-work mechanisms for DLTs suffer from high latencies of operations and enormous energy costs. Byzantine fault-tolerant (BFT) consensus protocols prove to be a potentially energy-efficient alternative to proof-of-work. However, current BFT protocols also present challenges that still limit their practical use in production systems. This research project addresses these challenges by (1) improving the scalability of BFT consensus protocols without reducing their resilience, (2) applying modelling approaches for making the expected performance and timing behaviour of these protocols more predictable, even under attacks, taking into consideration environmental conditions, and (3) supporting the design process for valid, automated testable BFT systems from specification to deployment in a blockchain infrastructure. The topic of scalability aims at finding practical solutions that take into account challenges such as recovery from major outages or upgrades, as well as reconfigurations at runtime. We also want to design a resilient communication layer that decouples the choice of a suitable communication topology from the actual BFT consensus protocol and thus reduces its complexity.This should be supported by the use of trusted hardware components. In addition, we want to investigate combinations of these concepts with suitable cryptographic primitives to further improve scalability. Using systematic modelling techniques, we want to be able to analyse the efficiency of scalable, complex BFT protocols (for example, in terms of throughput and latency of operations), already before deploying them in a real environment, based on knowledge of system size, computational power of nodes, and basic characteristics of the communication links. We also want to investigate robust countermeasures that help defending against targeted attacks in large-scale blockchain systems. The third objective is to support the systematic and valid implementation in a practical system, structured into a constructive, modular approach, in which a validatable BFT protocol is assembled based on smaller, validatable building blocks; the incorporation of automated test procedures based on a heuristic algorithm which makes the complex search space of misbehaviour in BFT systems more manageable; and a tool for automated deployment with accompanying benchmarking and stress testing in large-scale DLTs.

→ More information