Distributed Computing

Machine learning (ML) solutions are nowadays distributed, according to the so-called server/worker architecture. One server holds the model parameters while several workers train the model. Clearly, such architecture is prone to various types of ...$\frac{\mathrm{}}{\mathrm{}}$${}_{\mathit{}}\mathrm{}{}_{\mathit{}}\mathrm{}$$\frac{\mathrm{}}{\mathrm{}}$${}_{}\mathrm{}{}_{}$

The celebrated result of Fischer, Lynch and Paterson is the fundamental lower bound for asynchronous fault tolerant computation: any 1-crash resilient asynchronous agreement protocol must have some (possibly measure zero) probability of not ...$\mathrm{}$

Computing shortest paths from a single source is one of the central problems studied in the CONGEST model of distributed computing. After many years in which no algorithmic progress was made, Elkin [STOC ‘17] provided the first improvement over ...$\stackrel{}{}\sqrt{}{}^{\mathrm{}\mathrm{}}\sqrt{}{}^{\mathrm{}\mathrm{}}{}^{\mathrm{}\mathrm{}}$$\stackrel{}{\mathrm{}}\sqrt{}$$\mathrm{}\mathrm{}$$\stackrel{}{}\sqrt{}{}^{\mathrm{}\mathrm{}}$$\stackrel{}{}\sqrt{}{}^{\mathrm{}\mathrm{}}$

In modern operating systems and programming languages adapted to multicore computer architectures, parallelism is abstracted by the notion of execution threads. Multi-threaded systems have two major specificities: on the one part, new threads can ...