Related papers: When Load Rebalancing Does Not Work for Distribute…
In LLM serving, reusing the KV cache of prompts across requests is critical for reducing TTFT and serving costs. Cache-affinity scheduling, which co-locates requests with the same prompt prefix to maximize KV cache reuse, often conflicts…
This paper addresses the problem of efficiently storing and accessing massive data blocks in a large-scale distributed environment, while providing efficient fine-grain access to data subsets. This issue is crucial in the context of…
We consider a load balancing system consisting of $n$ single-server queues working in parallel, with heterogeneous service rates. Jobs arrive to a central dispatcher, which has to dispatch them to one of the queues immediately upon arrival.…
Anticipated rapid growth of large digital load, driven by artificial intelligence (AI) data centers, is poised to increase uncertainty and large fluctuations in consumption, threatening the stability, reliability, and security of the energy…
We propose a distributed algorithm for sparse signal recovery in sensor networks based on Iterative Hard Thresholding (IHT). Every agent has a set of measurements of a signal x, and the objective is for the agents to recover x from their…
In the traditional two-rack distributed storage system (DSS) model, due to the assumption that the storage capacity of each node is the same, the minimum bandwidth regenerating (MBR) point becomes infeasible. In this paper, we design a new…
Hadoop is an open source implementation of the MapReduce Framework in the realm of distributed processing. A Hadoop cluster is a unique type of computational cluster designed for storing and analyzing large data sets across cluster of…
Performance of standard processes over large distributed networks typically scales with the size of the network. For example, in planar topologies where nodes communicate with their natural neighbors, the scaling factor is $O(n)$, where $n$…
The emergence of THz (Terahertz) frequency wireless networks holds great potential for advancing various high-demand services, including Industrial Internet of Things (IIoT) applications. These use cases benefit significantly from the…
Reverse time migration (RTM) is a prominent technique in seismic imaging. Its resulting subsurface images are used in the industry to investigate with higher confidence the existence and the conditions of oil and gas reservoirs. Because of…
Our increasing reliance on the cloud has led to the emergence of scale-out workloads. These scale-out workloads are latency-sensitive as they are user driven. In order to meet strict latency constraints, they require massive computing…
Nowadays most of the cloud applications process large amount of data to provide the desired results. Data volumes to be processed by cloud applications are growing much faster than computing power. This growth demands new strategies for…
Modern applications demand high performance and cost efficient database management systems (DBMSs). Their workloads may be diverse, ranging from online transaction processing to analytics and decision support. The cloud infrastructure…
Distributed software systems that are designed to run over workstation machines within organisations are termed workstation-based. Workstation-based systems are characterised by dynamically changing sets of machines that are used primarily…
Introduction. Distributed data processing and storage systems require efficient methods to distribute keys across buckets. While simple and fast, the traditional modulo-based mapping is unstable when the number of buckets changes, leading…
A common approach to scaling transactional databases in practice is horizontal partitioning, which increases system scalability, high availability and self-manageability. Usu- ally it is very challenging to choose or design an optimal…
Internet-enabled things and devices operating in the physical world are increasingly integrated in modern distributed systems, supporting functionalities that require assurances that certain critical requirements are satisfied by the…
The problem of minimizing mean response time of generic jobs submitted to a heterogenous distributed computer systems is considered in this paper. A static load balancing strategy, in which decision of redistribution of loads does not…
The method of calculating a distributed system imbalance based on the calculation of node system load was proposed in the work. Calculation of node system load is carried out by calculating the average coefficient of utilization of CPU,…
We consider the data shuffling problem in a distributed learning system, in which a master node is connected to a set of worker nodes, via a shared link, in order to communicate a set of files to the worker nodes. The master node has access…