Related papers: A Smart Background Scheduler for Storage Systems
High-performance computing systems are complex machines whose behaviour is governed by the correct functioning of its many subsystems. Among these, the workload scheduler has a crucial impact on the timely execution of the jobs continuously…
Edge/fog computing, as a distributed computing paradigm, satisfies the low-latency requirements of ever-increasing number of IoT applications and has become the mainstream computing paradigm behind IoT applications. However, because large…
Although High Performance Computing (HPC) users understand basic resource requirements such as the number of CPUs and memory limits, internal infrastructural utilization data is exclusively leveraged by cluster operators, who use it to…
Many hardware structures in today's high-performance out-of-order processors do not scale in an efficient way. To address this, different solutions have been proposed that build execution schedules in an energy-efficient manner. Issue time…
Hosting diverse large language model workloads in a unified resource pool through co-location is cost-effective. For example, long-running chat services generally follow diurnal traffic patterns, which inspire co-location of batch jobs to…
Performing efficient resource provisioning is a fundamental aspect for any resource provider. Local Resource Management Systems (LRMS) have been used in data centers for decades in order to obtain the best usage of the resources, providing…
Kubernetes provides native autoscaling mechanisms, including the Horizontal Pod Autoscaler, Vertical Pod Autoscaler, and node-level autoscalers, to enable elastic resource management for cloud-native applications. However, production…
Large Language Model (LLM) workloads have distinct prefill and decode phases with different compute and memory requirements which should ideally be accounted for when scheduling input queries across different LLM instances in a cluster.…
GPU clusters have become essential for training and deploying modern AI systems, yet real deployments continue to report average utilization near 50%. This inefficiency is largely caused by fragmentation, heterogeneous workloads, and the…
The ever-increasing gap between compute and I/O performance in HPC platforms, together with the development of novel NVMe storage devices (NVRAM), led to the emergence of the burst buffer concept - an intermediate persistent storage layer…
Storage systems have not kept the same technology improvement rate as computing systems. As applications produce more and more data, I/O becomes the limiting factor for increasing application performance. I/O congestion caused by concurrent…
Low-latency online services have strict Service Level Objectives (SLOs) that require datacenter systems to support high throughput at microsecond-scale tail latency. Dataplane operating systems have been designed to scale up multi-core…
Task-based programming models are excellent tools to parallelize and seamlessly load balance an application workload. However, the integration of I/O intensive applications and task-based programming models is lacking. Typically, I/O…
Large Language Models (LLMs) such as GPT-4 and Llama have shown remarkable capabilities in a variety of software engineering tasks. Despite the advancements, their practical deployment faces challenges, including high financial costs, long…
Lightweight containers provide an efficient approach for deploying computation-intensive applications in network edge. The layered storage structure of container images can further reduce the deployment cost and container startup time.…
Data processing frameworks such as Apache Beam and Apache Spark are used for a wide range of applications, from logs analysis to data preparation for DNN training. It is thus unsurprising that there has been a large amount of work on…
This paper considers single-machine scheduling problems in which a given solution, i.e. an ordered set of jobs, has to be improved as much as possible by re-sequencing the jobs. The need for rescheduling may arise in different contexts,…
Modern latency-critical online services often rely on composing results from a large number of server components. Hence the tail latency (e.g. the 99th percentile of response time), rather than the average, of these components determines…
Dynamic offloading of Machine Learning (ML) model partitions across different resource orchestration services, such as Function-as-a-Service (FaaS) and Infrastructure-as-a-Service (IaaS), can balance processing and transmission delays while…
Modern large-scale computing systems distribute jobs into multiple smaller tasks which execute in parallel to accelerate job completion rates and reduce energy consumption. However, a common performance problem in such systems is dealing…